PowerExchange Reference Manual - Informatica · The site contains product information, user group information, newsletters, access to the Informatica customer support case management

Informatica® PowerExchange®

(Version 8.6.1 HotFix 10)

Reference Manual

Informatica PowerExchange Reference Manual

Version 8.6.1 February 2010

Copyright (c) 1998–2010 Informatica Corporation. All rights reserved.

This software and documentation contain proprietary information of Informatica Corporation and are provided under a license agreement containing restrictions on use and disclosure and are also protected by copyright law. Reverse engineering of the software is prohibited. No part of this document may be reproduced or transmitted in any form, by any means (electronic, photocopying, recording or otherwise) without prior consent of Informatica Corporation. This Software may be protected by U.S. and/or international Patents and other Patents Pending.

Use, duplication, or disclosure of the Software by the U.S. Government is subject to the restrictions set forth in the applicable software license agreement and as provided in DFARS 227.7202-1(a) and 227.7702-3(a) (1995), DFARS 252.227-7013(c)(1)(ii) (OCT 1988), FAR 12.212(a) (1995), FAR 52.227-19, or FAR 52.227-14 (ALT III), as applicable.

The information in this product or documentation is subject to change without notice. If you find any problems in this product or documentation, please report them to us in writing.

Informatica, PowerCenter, PowerCenterRT, PowerCenter Connect, PowerCenter Data Analyzer, PowerExchange, PowerMart, Metadata Manager, Informatica Data Quality, Informatica Data Explorer, Informatica B2B Data Exchange and Informatica On Demand are trademarks or registered trademarks of Informatica Corporation in the United States and in jurisdictions throughout the world. All other company and product names may be trade names or trademarks of their respective owners.

Copyright, Byte Designs Ltd. All rights reserved.

This product includes ICU software which is copyright (c) 1995-2003 International Business Machines Corporation and others. All rights reserved. Permissions and limitations regarding this software are subject to terms available at http://www-306.ibm.com/software/globalization/icu/license.jsp.

The product includes the zlib library copyright (c) 1995-2005 Jean-loup Gailly and Mark Adler.

DISCLAIMER: Informatica Corporation provides this documentation “as is” without warranty of any kind, either express or implied, including, but not limited to, the implied warranties of non-infringement, merchantability, or use for a particular purpose. Informatica Corporation does not warrant that this software or documentation is error free. The information provided in this software or documentation may include technical inaccuracies or typographical errors. The information in this software and documentation is subject to change at any time without notice.

Part Number: PWX-REF-861-0007

Table of Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Informatica Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Informatica Customer Portal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Informatica Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Informatica Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Informatica How-To Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Informatica Knowledge Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Informatica Multimedia Knowledge Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Informatica Global Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Chapter 1: PowerExchange Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

PowerExchange Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Chapter 2: DBMOVER Configuration File Parameters . . . . . . . . . . . . . . . . . . . . . . . . . 3

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Syntax Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

DBMOVER Parameter Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

PowerExchange Message Log Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

PowerExchange Log File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

PowerExchange Alternative Log Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Consumer API (CAPI) Connection Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Multiple CAPI Connections and Overrides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Source-Specific CAPI_CONNECTION Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

pwxcmd Command Configuration Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

SVCNODE Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

CMDNODE Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Example DBMOVER Configuration Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Chapter 3: Netport Jobs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Netport Jobs Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Configuring for Netport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Netport Jobs Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

PowerExchange Substitution Variables for Netport Jobs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Netport Jobs and Offload Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Netport Jobs and DB2 Bulk Data Loads. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Using Netport Jobs for Generation Data Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

The Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

The Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Setting Up Netport Jobs for GDGs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Table of Contents iii

Chapter 4: Message Destination Override . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

DTLMSGO File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Syntax Rules for DTLMSGO Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Syntax for DTLMSGO Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Required Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

Optional Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

DTLMSGO File Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

Chapter 5: Statistics Logging Including SMF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

Configuring for SMF on z/OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Configuring for File Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Record Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Standard SMF Header with Sub Type Included . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

PowerExchange Header/Descriptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Extended Section Descriptor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

General Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Extended Section for PowerExchange Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Extended Section for PowerExchange Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

Extended Section for Data Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Function/Access Method Section (DB2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Chapter 6: PowerExchange Security. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

PowerExchange Security Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

SECURITY Statement in the DBMOVER Configuration File . . . . . . . . . . . . . . . . . . . . . . . . 86

i5/OS Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Resource Access Requirements for i5/OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Linux, UNIX, and Windows Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

User Authentication for the pwxcmd Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

Authorization for Running pwxcmd Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

MVS Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Resource Access Requirements on MVS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Security for Capture Registrations and Extraction Maps . . . . . . . . . . . . . . . . . . . . . . . . 95

Security for Data Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Security for PowerExchange Listener Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Adabas Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Datacom Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

DB2 for z/OS Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

IMS Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

PowerExchange Selective Sign-on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

PowerExchange Sign-on File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

PowerExchange Sign-on File Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

PowerExchange Sign-on File Statements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

PowerExchange Sign-on File Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

iv Table of Contents

Chapter 7: Secure Sockets Layer Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

SSL Security on a PowerExchange Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Application Transparent - Transport Layer Security (AT-TLS) . . . . . . . . . . . . . . . . . . . . 106

Configuring SSL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Configuring SSL Client and Server on MVS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Configuring SSL Server on Windows, UNIX, or Linux . . . . . . . . . . . . . . . . . . . . . . . . . 107

Configuring the SSL Client on Windows, UNIX, or Linux . . . . . . . . . . . . . . . . . . . . . . 107

Running Secure and Non-secure Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

SSL Authentication Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Sample Certificate Authority List (CALIST) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Private Key . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Certificate Authority List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

AT-TLS Policy File for PowerExchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Chapter 8: Encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Modes of Encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

PowerExchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

DES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Triple DES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

RC2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Methods of Encrypting Data in PowerExchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Setting Defaults in the DBMOVER Configuration File . . . . . . . . . . . . . . . . . . . . . . . . 112

Chapter 9: Nonrelational SQL Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Supported Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Basic Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

LIKE Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

BETWEEN Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

NULL Comparison . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Allowed Keywords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Accessing Data with Column Names the Same as SQL Keywords . . . . . . . . . . . . . . . . . 116

Using SQL with User-defined Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

IMS Call Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Metadata Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

DTLDESCRIBE TABLES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

DTLDESCRIBE PROCEDURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

DTLDESCRIBE PROCEDURECOLUMNS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

DTLDESCRIBE COLUMNS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

DTLDESCRIBE RECORDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

DTLDESCRIBE PKEYS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

DTLDESCRIBE FKEYS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

DTLDESCRIBE SCHEMAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

DTLDESCRIBE Qualifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

DB2 for i5/OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Table of Contents v

DB2 for Linux, UNIX, and Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

DB2 for z/OS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Informix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

NRDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

NRDB2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Oracle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Sybase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Chapter 10: National Language Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

Default Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

General Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Code Pages for Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

DB2 for i5/OS Code Page Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

DB2 for z/OS Code Page Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

Oracle Code Page Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

User-Defined Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

ICU-Compatible Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

PowerExchange Static Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

ICUCHECK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Report 1. Power Exchange Control Table for Simple Code Pages . . . . . . . . . . . . . . . . . . 141

Report 2. PowerExchange Control Table for ICU Code Pages . . . . . . . . . . . . . . . . . . . . 142

Report 3. Comparing Name and Character Size Information with ICU . . . . . . . . . . . . . 143

Report 4. Comparing ICU Code Page Information with PowerExchange . . . . . . . . . . . . 144

Report 5. PowerExchange Code Page Names and Aliases . . . . . . . . . . . . . . . . . . . . . . . . 145

Report 6. PMlocale Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Report 7. Control Counts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Report 8. Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Code Pages Shipped with PowerExchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

PowerExchange Static Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Code Pages Holding All Unicode Characters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Code Pages Based on ASCII Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

Specialized ICU Code Pages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Handling Conversion Errors and Special Situations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Conversion Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Data in a Code Page That the Database Does Not Expect . . . . . . . . . . . . . . . . . . . . . . . 156

Chapter 11: Using the PowerExchange ODBC Driver. . . . . . . . . . . . . . . . . . . . . . . . 157

PowerExchange ODBC Driver Data Source Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

General PowerExchange ODBC Driver Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

PowerExchange ODBC Driver Parameters for DB2 for z/OS Bulk Load Processing . . . . 163

Creating an ODBC Data Source on Windows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

PowerExchange Data Source Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

General Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

DB2/S390 Bulk Load Properties Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

vi Table of Contents

AS/400 Properties Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

EMR Properties Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

CAPX Properties Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

CAPXRT Properties Tab . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

Accessing Multibyte Metadata with the PowerExchange ODBC Driver . . . . . . . . . . . . . 169

Creating an ODBC Data Source on UNIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Example PowerExchange Data Source Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Accessing Multibyte Metadata with ODBC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

SQL Escape Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

SQL Escape Sequences with PowerCenter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

SQL Escape Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

{DTLADAPREFETCH=Y|N} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLADAPWD= }. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLAPP= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLCONFWRITE=N|Y|T} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLCONNOVR= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLDATAPWD= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLDB2DEGREE= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

{DTLDSN= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

{DTLEVENTTB= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

{DTLIMTYPE=BA|AI|TU} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

{DTLIMGOV=Y|N} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

{DTLINSMODE=LOAD|UPDATE} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLJRNL= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLLIBRARYLIST= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLNOUPDATECDEP=Y|N} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLORACOLL= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLORACONN= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLORAINST= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174

{DTLORASCHEMA= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLOVRDBF= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLREJECTFILE= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLSESSID=} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLSTOPAFTER=} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLSTRIPORDERBY=Y|N} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175

{DTLTIMEOUT= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

{DTLUDBDB= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

{DTLUPDELSEINS=Y} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

{DTLWORKERS= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

{DTLXTRASCHEMA= } . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

{DTLXTYPE=RS|SL} . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

Chapter 12: Datatypes and Conversion Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Datatype Conversions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

ODBC Datatype Equivalency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178

Table of Contents vii

Appendix A: PowerExchange Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

Glossary of Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195

viii Table of Contents

Preface

This manual provides reference information for subjects related to configuring and using PowerExchange®. The manual is not intended to be read cover-to-cover. Use it to look up information about a particular subject.

The manual contains information that you need to fully configure and tune your PowerExchange system. For example, it includes information about configuring product security and describes each parameter in the DBMOVER configuration file. The manual also contains information that you need to use PowerExchange in certain environments. For example, if you have nonrelational data sources, the manual describes the SQL that PowerExchange supports to access that data.

Use this manual in conjunction with the PowerExchange Installation Guide, PowerExchange Bulk Data Movement Guide, and the platform-specific guides to gain a complete understanding of the product.

This manual applies to the following PowerExchange products:

♦ PowerExchange for Adabas®

♦ PowerExchange for CA Datacom®

♦ PowerExchange for CA IDMS™

♦ PowerExchange for DB2® for i5/OS®

♦ PowerExchange for DB2 for Linux®, UNIX®, and Windows®

♦ PowerExchange for DB2 for z/OS®

♦ PowerExchange for IMS™

♦ PowerExchange for Oracle®

♦ PowerExchange for SQL Server®

♦ PowerExchange for VSAM

Informatica Resources

Informatica Customer PortalAs an Informatica customer, you can access the Informatica Customer Portal site at http://my.informatica.com. The site contains product information, user group information, newsletters, access to the Informatica customer support case management system (ATLAS), the Informatica How-To Library, the Informatica Knowledge Base, the Informatica Multimedia Knowledge Base, Informatica Documentation Center, and access to the Informatica user community.

ix

Informatica DocumentationThe Informatica Documentation team takes every effort to create accurate, usable documentation. If you have questions, comments, or ideas about this documentation, contact the Informatica Documentation team through email at [email protected]. We will use your feedback to improve our documentation. Let us know if we can contact you regarding your comments.

The Documentation team updates documentation as needed. To get the latest documentation for your product, navigate to the Informatica Documentation Center from http://my.informatica.com.

Informatica Web SiteYou can access the Informatica corporate web site at http://www.informatica.com. The site contains information about Informatica, its background, upcoming events, and sales offices. You will also find product and partner information. The services area of the site includes important information about technical support, training and education, and implementation services.

Informatica How-To LibraryAs an Informatica customer, you can access the Informatica How-To Library at http://my.informatica.com. The How-To Library is a collection of resources to help you learn more about Informatica products and features. It includes articles and interactive demonstrations that provide solutions to common problems, compare features and behaviors, and guide you through performing specific real-world tasks.

Informatica Knowledge BaseAs an Informatica customer, you can access the Informatica Knowledge Base at http://my.informatica.com. Use the Knowledge Base to search for documented solutions to known technical issues about Informatica products. You can also find answers to frequently asked questions, technical white papers, and technical tips. If you have questions, comments, or ideas about the Knowledge Base, contact the Informatica Knowledge Base team through email at [email protected].

Informatica Multimedia Knowledge BaseAs an Informatica customer, you can access the Informatica Multimedia Knowledge Base at http://my.informatica.com. The Multimedia Knowledge Base is a collection of instructional multimedia files that help you learn about common concepts and guide you through performing specific tasks. If you have questions, comments, or ideas about the Multimedia Knowledge Base, contact the Informatica Knowledge Base team through email at [email protected].

Informatica Global Customer SupportThere are many ways to access Informatica Global Customer Support. You can contact a Customer Support Center through telephone, email, or the WebSupport Service.

Use the following email addresses to contact Informatica Global Customer Support:

♦ [email protected] for technical inquiries

♦ [email protected] for general customer service requests

WebSupport requires a user name and password. You can request a user name and password at http://my.informatica.com.

x Preface

Use the following telephone numbers to contact Informatica Global Customer Support:

North America / South America Europe / Middle East / Africa Asia / Australia

Informatica Corporation Headquarters100 Cardinal WayRedwood City, California 94063United States

Toll Free +1 877 463 2435

Standard RateBrazil: +55 11 3523 7761 Mexico: +52 55 1168 9763 United States: +1 650 385 5800

Informatica Software Ltd.6 Waltham ParkWaltham Road, White WalthamMaidenhead, BerkshireSL6 3TNUnited Kingdom

Toll Free 00 800 4632 4357

Standard RateBelgium: +32 15 281 702France: +33 1 41 38 92 26Germany: +49 1805 702 702Netherlands: +31 306 022 797Spain and Portugal: +34 93 480 3760United Kingdom: +44 1628 511 445

Informatica Business Solutions Pvt. Ltd.Diamond DistrictTower B, 3rd Floor150 Airport RoadBangalore 560 008India

Toll Free Australia: 1 800 151 830Singapore: 001 800 4632 4357

Standard RateIndia: +91 80 4112 5738

Preface xi

xii Preface

C H A P T E R 1

PowerExchange Overview

This chapter includes the following topics:

♦ Overview, 1

♦ PowerExchange Components, 2

Overview

PowerExchange can perform the following functions to synchronize a data source and data target:

♦ Bulk data movement to materialize or entirely refresh a data target.

♦ Change data capture to keep a data source and data target synchronized.

PowerExchange bulk data movement can move large amounts of data between different platforms efficiently. However, frequent bulk data movement operations can be costly in terms of space, time, and staff resources.

With PowerExchange change data capture (CDC), you can capture only the changes that are made to a data source. PowerExchange CDC has minimal impact on the performance and availability of the source database, tables, and files. PowerExchange provides two methods of change data capture:

♦ Synchronous change data capture, which occurs in real time by integrating into the transaction performing the change

♦ Asynchronous change data capture, also called log-based changed data capture, which captures the changes from the source database or source relational database logs

PowerExchange works with PowerCenter to enable you move bulk data and change data to a variety of data targets.

This manual provides reference information for customizing PowerExchange for your environment. After installing the product, review the topics on the DBMOVER configuration file parameters and product security options. Also, review any of the other topics that might pertain to your specific environment and data replication requirements.

1

PowerExchange Components

PowerExchange uses some or all of the following components to perform its functions, depending on the platform that you are using:

PowerExchange Navigator The graphical user interface from which you define and manage data maps, capture registrations, and extraction maps for the data sources from which you want to extract bulk data or capture change data. PowerExchange uses these definitions to determine the data sources to process. Capture registrations and extraction maps pertain to change data capture (CDC) only.

PowerExchange ListenerManages data maps for nonrelational files and DB2 tables, and capture registrations and extraction maps for all data sources. The PowerExchange Listener maintains these definitions in the following files:

♦ Data maps in the DATAMAPS file

♦ Capture registrations in the CCT file

♦ Extraction maps in the DTLCAMAP or CAMAPS file

The PowerExchange Listener also handles extraction requests for bulk data and change data.

If a data source or data target is remote from the system on which you are using PowerExchange, you must also run a PowerExchange Listener on the remote data source or target to communicate with PowerExchange.

PowerExchange AgentOn an MVS system, provides capture registration information to the following ECCRs during CDC:

♦ DB2

♦ IMS synchronous

♦ Batch VSAM

♦ CICS/VSAM

Other ECCRs read capture registration information directly from the CCT data set. For all of the ECCRs, the PowerExchange Agent verifies the capture registration information.

The PowerExchange Agent also manages global queues and data flow among various PowerExchange CDC components.

PowerExchange Logger for Linux, UNIX, and WindowsOn a Linux, UNIX, or Windows machine, optionally writes change data from the change stream to PowerExchange Logger log files. PowerExchange then extracts the change data from the PowerExchange Logger log files rather than from the source database or log files. A PowerExchange Logger process can operate in either continuous mode or batch mode.

PowerExchange Logger for MVSOn an MVS system, receives captured change data from the PowerExchange Environmental Change Capture Routines (ECCR) that are connected to it and stores the change data in log data sets. The change data is then available for real-time extractions or PowerExchange Condense jobs.

PowerExchange Environmental Change Capture Routine (ECCR) On an MVS system, captures change data from a data source and passes the captured changes to the PowerExchange Logger for recording. PowerExchange provides an ECCR for each type of data source. Depending on the source type, the ECCR captures changes synchronously as they are made or asynchronously from database logs.

PowerExchange CondenseOn MVS or i5/OS, optionally creates condense files that contain a condensed version of the change data in the change stream. If you use this optional feature, PowerExchange extracts change data from condense files rather than from log files. You can run multiple Condense jobs at a time.

2 Chapter 1: PowerExchange Overview

C H A P T E R 2

DBMOVER Configuration File Parameters


♦ Overview, 3

♦ Syntax Rules, 4

♦ DBMOVER Parameter Descriptions, 4

♦ PowerExchange Message Log Options, 36

♦ Consumer API (CAPI) Connection Statements, 42

♦ ICU-specific Configuration Parameters, 62

♦ pwxcmd Command Configuration Statements, 64

♦ Example DBMOVER Configuration Files, 66

Overview

The PowerExchange DBMOVER configuration file contains parameters that control PowerExchange operations and communications. Define a DBMOVER configuration file on each system where PowerExchange software is installed.

PowerExchange provides a sample DBMOVER configuration file for each supported platform. The information in a sample file is sufficient to establish basic communication. Usually, you must edit the sample file to customize it, for example, to add database instances, add PowerExchange nodes with which to communicate, configure CDC, and specify localization parameters. For more information, see “Example DBMOVER Configuration Files” on page 66.

3

Syntax Rules

Use the following rules and guidelines when you configure the DBMOVER configuration file:

♦ Every parameter starts on a new line.

♦ If a parameter has more than one option, include them in parentheses and separate them by a comma (,) or a space. For example:

NODE=(node_name,TCPIP,host_name,2480)

♦ If you need to include a space in a parameter value such as a Windows path, the complete parameter value must be enclosed in double quotation marks, for example:

"aaa bbbb\cccc"

Note: Use straight quotes ("). Do not use curly quotes (“).

DBMOVER Parameter Descriptions

The following table describes all parameters that you can include in a DBMOVER configuration file:

Parameter DefaultParameter Options

Description

ABEND_SW None Y Installs the PowerExchange exit handlers to trap and handle program abends. Default behavior is to let the standard C program handlers handle the abend and produce a core file.MVS only.

4 Chapter 2: DBMOVER Configuration File Parameters

ADABASCODEPAGE Specifies the single-byte and multiple-byte code pages to use for an Adabas database. This parameter is optional. Usually, this parameter is specified only for databases that have WIDECHAR fields that require a code page other than the default code page or the code page that is set in the CODEPAGE parameter.

Syntax:ADABASCODEPAGE=(database_id, sb_codepage,mb_codepage)

Where:- database_id. The identifier for the

Adabas database or file to which the code page or code pages apply. This value is required. You can enter 0 to indicate the Adabas default database.

- sb_codepage. The name of a single-byte code page.

- mb_codepage. The name of a multiple-byte code page.

If the database contains WIDECHAR fields, you should enter a multiple-byte code page.

You can enter up to 20 ADABASCODEPAGE statements in the DBMOVER configuration file.

You can override the code pages that are specified in this parameter when you define a data map. In the data map, you can specify a code page for a specific source file and a Wide-character code page that applies only to WIDECHAR fields in the file. Additionally, you can specify code pages for specific fields. A field-level code page overrides a data map code page, and a data map code page overrides any code page that is specified in this parameter or any other DBMOVER code page parameter.

ADABAS_DEFAULT_DBID None DBID Specified as a DBID in the format ADABAS_DEFAULT_DBID=n, where n is the default DBID. If the DBID in the data map is 0, and if no override is specified in the Override File Name in the Database Connection or ODBC parameters, this default DBID is used. Setting the DBID in the data map to 0 and using this parameter can help when migrating data maps from one environment to another.


Description

DBMOVER Parameter Descriptions 5

ADABAS_PREFIX DTL0 1 through 4 alphanumeric characters

Specifies the prefix that PowerExchange uses to construct an ET-ID for accessing Adabas files in the PowerExchange Listener. Each ET-ID is composed of this prefix followed by a four-digit task number.Use this parameter if you run multiple PowerExchange Listeners that need to access the same Adabas database. You must specify a different prefix for each PowerExchange Listener so that PowerExchange can generate a unique ET-ID value for each one. If the ET-IDs are not unique, one or more of the PowerExchange Listeners might not be able to access the Adabas database. In this case, the PowerCenter session fails.

ADA_L3_ALLOW N Y or N Enables the use of Adabas L3 logical reads but, unlike ADAOPT, does not cause PowerExchange to check with Adabas that the version of Adabas software running is V7 or greater. Use this parameter when user modifications to Adabas prevent PowerExchange from determining the version of software installed.

ADAOPT N Y or N Adabas optimization parameter. Valid values:- N. Do not use keys for Adabas Reads.- Y. Use descriptors, superdescriptors and

subdescriptors, whenever possiblePowerExchange interrogates Adabas to ensure that Adabas version 7 or later is running. Version 7 is the minimum version for this access method to work.

ADAOPTM B A or B Adabas optimization:- A. Use keys whenever possible.- B. Only use keys when a user has given

both start and end values.

ADAPREFETCH N Y or N Enables PowerExchange to use the ADABAS Pre-Fetch functionality to enhance performance.

ADASTATS N Y or N Extracts statistics describing ADABAS data retrieval. If statistics is enabled, the resulting information is written to the PowerExchange log file. The following information is written:- Format Buffer being used.- Whether Pre-Fetch is being used.- Type of read being used for the request:

L2 or L3 (physical or logical).- If a logical read is being used, the key

being used and key values.- Number of records that have been read.


Description


ADAUSER DTL00011 JOBNAME Specify this parameter with the value JOBNAME if you run multiple Netport jobs for accessing Adabas data on z/OS and want to use the Netport job name of each as the unique user ID in Adabas.

When multiple jobs run under the same user ID, such as the default, the jobs fail with Adabas Response Code 48 Sub Code 8 and PowerExchange message PWX-00416. Use this parameter to prevent this problem.

If you do not specify a value, PowerExchange uses either the default value of DTL00011 or, if specified, the ADABAS_PREFIX value appended with 0011.

This statement pertains to Netport jobs only. If you access Adabas data through a PowerExchange Listener, this statement is ignored.

APPBUFSIZE 128000 34816 through 1048576

Defines the maximum data buffer size in bytes. When data rows in excess of this amount have been read, the buffer is sent to the partner platform. Must be greater than the maximum size of a single row that will be sent.

AS400EVENTMSGQ None An event message queue on an i5/OS system to which PowerExchange writes a message when AS4JRNEXIT=Y or ALWCLRPFM=Y is specified on the AS4J CAPI_CONNECTION statement.

Enter the message queue in the following format:library/queue_name

- If AS4JRNEXIT=Y, PowerExchange writes the message DTL3001 for each journal receiver that is processed by a change data extraction. PowerExchange writes the message when it reads the first journal entry for the next journal receiver on the chain. If multiple change data extractions successfully process the same journal receiver, the message queue contains multiple DTL3001 messages for that journal receiver.

- If ALWCLRPFM=Y, PowerExchange writes the message DTL3002 to the queue if a journal entry for a CLRPFM command is encountered. Note that you should not specify the ALWCLRPFM parameter unless instructed to do so by Informatica Global Customer Support.

For more information about using event message queues, see the PowerExchange CDC Guide for i5/OS.


Description


BS None 8 through 32760

Default BLKSIZE to use value for dynamic allocation.MVS only.

CAPI_CONN_NAME None Default CAPI_CONNECTION statement, across all source types.This parameter is optional. For more information, see “Consumer API (CAPI) Connection Statements” on page 42.

CAPI_CONNECTION A named set of parameters that the PowerExchange Consumer API (CAPI) uses to connect to the change stream and control extraction processing. The CAPI_CONNECTION type depends on the data source type and extraction mode. You can define up to eight CAPI_CONNECTION statements in a DBMOVER configuration file for the same source type or for different source types. To indicate a default CAPI_CONNECTION, you can use the CAPI_SRC_DFLT or CAPI_CONN_NAME parameter.For more information, see “Consumer API (CAPI) Connection Statements” on page 42.

CAPI_SRC_DFLT None Default CAPI_CONNECTION statement for a source type such as DB2 or Oracle. If you defined multiple CAPI_CONNECTION statements for a source type, you can use this parameter to identify one of them as the default.

Syntax is:CAPI_SRC_DFLT=(source_type, capi_connection_name)

Where:- source_type. One of the following source

database types: ADA, AS4, DB2, DCM, IDL, IDM, IML, IMS, IXL, MSS, ORA, TIM, UDB, VSAM, or VSM.

- capi_connection_name. The unique name of the CAPI_CONNECTION statement that you want to use as the default.

You can specify a single CAPI_SRC_DFLT statement for each source type. This parameter is optional.

You can override the default CAPI_CONNECTION in multiple ways. For more information, see “Consumer API (CAPI) Connection Statements” on page 42.


Description


CAPT_PATH None Path to the local directory that stores the following files for CDC:- CCT file. Contains capture registrations.- CDEP file. Contains application names

for PowerCenter extractions that use ODBC connections, if any.

- CDCT file. Contains information about PowerExchange Logger for Linux, UNIX, and Windows log files, if you use the PowerExchange Logger.

This directory can be a directory that you created specifically for these files or another existing directory. Informatica recommends that you use a unique directory name to separate these CDC objects from the PowerExchange code. This practice makes migrating to a new PowerExchange version easier.

Applies to the following data sources: Microsoft SQL Server, Oracle, and DB2 for Linux, UNIX, and Windows. This parameter is required for CDC.

CAPT_XTRA None Path to the local directory that stores extraction maps.

This directory can be a directory that you created specifically for extraction maps or another existing directory.

Applies to the following data sources: Microsoft SQL Server, Oracle, and DB2 for Linux, UNIX, and Windows.Required for CDC.

CMDNODE On Linux, UNIX, and Windows, use the CMDNODE statement to specify a PowerExchange service to which you want to send pwxcmd commands.

For more information, see “pwxcmd Command Configuration Statements” on page 64.

CONSOLE_CODEPAGE Writes console output in the control code page

Code page identifier

Code page in which message text is displayed by PowerExchange console applications.

For example, specify the console code page in the following situations where the operating system locale is defined as using Japanese characters:- On Windows, the operating system uses

the OS locale derived from the user internationalization settings. Specify CONSOLE_CODEPAGE=CP943.

- On UNIX and Linux, the console emulator assumes the data to be in code page UTF-8. Specify CONSOLE_CODEPAGE=UTF-8.


Description


CODEPAGE By platform. See “National Language Support” on page 131.

Enables code page separate overrides for operating system information, data and any literals sent in SQL statements. This parameter does not affect the PowerExchange Navigator, which always works in UTF-8. If this parameter is not set, PowerExchange uses the platform default (i5/OS IBM37, MVS IBM1047, Windows/UNIX ISO8859)Format:CODEPAGE=(ctrl_cp,data_cp, SQL_cp)

Where:- ctrl_cp. The control code page for

operating system interfaces (covers program names, file names, and so on).

- data_cp. Used to override the code page that data arrives in (for example, on Windows, data from MVS).

- SQL_cp. The code page of any literals within an SQL statement.

COLON : Single character

Used by the TXT time and timestamp routines as the separator character, for example:HH:MM:SS

COMMA , Single character

Parameter list separator.

COMPRESS Y Y or N If Y is specified, all data movements accessed by way of a PowerExchange Listener are compressed.

CONSOLE_TRACE N Y or N If Y is specified, all tracing is displayed on the MVS console device or i5/OS QPRINT output.


Description


CONVCHAR NULLandERROR

Defines the behavior for handling field conversion errors.If there is a conversion error in a CHAR or VARCHAR field, the character specified by CONVCHAR is used to fill the field. You can specify options to decide how the field is filled.CONVCHAR=(char, [{NULL|REPLACE}], [maxlen], [{NOERROR|ERROR}])

Where:- char. Replacement character.- NULL. (Default) Set the field to NULL if it

is nullable.- REPLACE. Always fill the field with the

replacement character.- maxlen. Maximum number of characters

to fill.- ERROR. (Default) Fields with conversion

errors are marked as such. Displays the error message PWX-000142 (conversion error) in the PowerExchange log.

- NOERROR. Fields with conversion errors are treated as if no error occurred; the replaced data appears in the field without returning an error. The error message PWX-000142 (conversion error) still appears in the PowerExchange log.

CPX_DIR The name of a library that is used on the i5/OS system to store the extraction maps that are used by change data capture. This value is written to the DBMOVER member during installation. It is written during the step for creating a PowerExchange environment by the CRTPWXENV command. For more information about CRTPWXENV, see the PowerExchange CDC Guide for i5/OS.Change CPX_DIR only under guidance of Informatica Global Customer Support. This parameter applies to i5/OS only.

CREDENTIALS_CASE D D, S, or A Specifies how PowerExchange interprets mixed case in login user IDs and passwords.Set to one of the following values: - D. Default if not specified in the

DBMOVER configuration file. This maintains the default behavior of PowerExchange by converting all user IDs and passwords to uppercase.

- S. The user ID and password case is specified by the customer. Exceptions to this are where particular systems (z/OS or i5/OS) do not allow a mixed case user ID.

- A. PowerExchange checks if the z/OS or i5/OS system is enabled to handle mixed-case passwords and then automatically capitalizes passwords, as appropriate.


Description


DATAMAP_SERVER None Listener Node Points to a PowerExchange Listener node where the PowerExchange Listener will access the data map file. Syntax:DATAMAP_SERVER=nodename

Where nodename is another PowerExchange Listener node in the node list. Message DTL-02805 indicates the data map server to be used in the associated PowerExchange Listener when it is started. In a normal configuration this parameter is not used in the configuration file of the PowerExchange Listener which runs as the data map server. The exception would be if configuring an environment when the first PowerExchange Listener to access the data map file would become the data map server. This would not be the usual configuration, but could be created as follows:Two or more PowerExchange Listeners can be set up with a DATAMAP_SERVER statement pointing to each other. Where two or more PowerExchange Listeners are set up using the same data map file, but each pointing to another PowerExchange Listener, the first PowerExchange Listener to start accesses the data map file and acquires that data map server status. The other PowerExchange Listeners then request data map access from that PowerExchange Listener.MVS only.

DATERANGE DATERANGE=(yyyy,yyyy)

Specifies a different date range. For example:DATERANGE=(1200,2800)

By default, PowerExchange can handle dates from 1800 through 2200.

DB2_BIN_AS_CHAR N Y or N Indicates if “FOR BIT DATA” CHAR and VARCHAR columns in DB2 for i5/OS are treated as BINARY. Valid values are:- N. These columns are described as

BINARY. Therefore, no code page translation is performed.

- Y. These columns are described as CHAR and therefore undergo code page translation.

Use this parameter in conjunction with DB2_BIN_CODEPAGE to override invalid 65535 CCSIDs assigned to character data by some packages.Default is N.This parameter applies to i5/OS only.


Description


DB2_BIN_CODEPAGE DB2 CCSIDs Defines the code page to be used if the DB2_BIN_AS_CHAR parameter is set to Y.Syntax:DB2_BIN_CODEPAGE= (sbcs_ccsid,dbcs_ccsid)

Where:- sbcs_ccsid. Single-byte character set

CCSID- dbcs_ccsid. Double-byte character set

CCSID.Use this parameter in conjunction with DB2_BIN_AS_CHAR to override invalid 65535 CCSIDs assigned to character data by some packages.For example:DB2_BIN_CODEPAGE=(8482,1390)

This parameter applies to i5/OS only.

DB2_ERRORFILE Character stringpath_filename

Specifies the path and file name of the fault tolerance error action files.For more information, see the PowerExchange Bulk Data Movement Guide.


Description


DB2CODEPAGE DB2 default Specifies the CCSID information for the DB2 subsystem. Syntax:DB2CODEPAGE=db2_subsystem, EBCDIC_CCSID=(sbcs_ccsid, dbcs_ccsid,mixed_ccsid), ASCII_CCSID=(sbcs_ccsid, dbcs_ccsid,mixed_ccsid), UNICODE_CCSID=(sbcs_ccsid, dbcs_ccsid,mixed_ccsid), [MIXED={Y|N}])

Where:- db2_ssid. The DB2 subsystem identifier.- sbcs_ccsid. The single-byte character

set.- dbcs_ccsid. The double-byte character

set.- mixed_ccsid. The mixed-character set

that contains both single- and double-byte characters.

The values that you specify must be the same as those specified for the DB2 subsystem in the DB2 installation panel DSNTIPF or in the equivalent specifications in the DSNHDECP module. The DB2CODEPAGE parameters are equivalent to the following DSNTIPF entries: - ASCII CCSID - EBCDIC CCSID- UNICODE CCSID - MIXED DATA

You must include the subsystem ID and EBCDIC_CCSID parameter. The ASCII_CCSID and UNICODE_CCSID parameters are optional. Enter 65534 for any CCSID parameter if you do not want to specify a CCSID for DB2.

Optionally, include the MIXED parameter to indicate whether ASCII and EBCDIC character strings can contain a mixture of SBCS and DBCS characters. Enter Y for yes or N for no. Default is N if MIXED is not specified. Example:DB2CODEPAGE=(D71G,EBCDIC_CCSID=(290,300,930),ASCII_CCSID=(1041,301,942),UNICODE_CCSID=(367,1200,1208),MIXED=Y)

If PowerExchange connects to a DB2 subsystem that does not have a DB2CODEPAGE statement, PowerExchange uses the default code page of the PowerExchange Listener.To view tables of CCSIDs, see the IBM DB2 for z/OS Installation Guide.MVS only.


Description


DB2DEF_ENCODING E - E. EBCDIC.- A. ASCII.- U. Unicode.

Specifies the default encoding scheme for DB2 data. Ensure that the encoding scheme for a particular database is the same as the DEF ENCODING SCHEME that is specified in the DB2 - APPLICATION PROGRAMMING DEFAULTS PANEL, DSNTIPF. Default is E if this parameter is not specified.MVS only.

DB2ID Character string

DB2 access module that PowerExchange uses to access a particular DB2 subsystem. The access module depends on the DB2 version of the subsystem. If you use different DB2 versions at your installation, ensure that you specify the correct access module for each subsystem that PowerExchange accesses.

Syntax:DB2ID=(ssid,plan_id,module)

Where:- ssid. The DB2 subsystem ID.- plan_id. The DB2 plan identifier.- module. The access method module

name.

You can specify the following access method module names: - default. Specifies the default access

method, which is DTLAMDB2.- DTLAMDB2. Uses single-row SQL

statements to read and write data.- DTLAMV8F. For DB2 Version 8

new-function mode or later, uses multiple-row FETCH and INSERT statements.

If you do not specify a module, DTLAMDB2 is used.

If you do not include a plan identifier, the value in the DB2PLAN parameter is used.

You can specify up to 25 DB2ID statements in a DBMOVER configuration member.

Note: You must APF-authorize the PowerExchange LOADLIB library to use multiple-row FETCH and INSERT statements.MVS only.

DB2PLAN Character string

DB2 plan name for DB2 bulk data processing. This plan name is the same as the plan name that you specified in the XIDDB210 bind job during PowerExchange installation.Default is PWXBKvrm, where vrm is the PowerExchange version, release, and modification level, for example, 850.MVS only.


Description


DECPOINT . Single character

Decimal point character to use in fields that contain non-integer numbers. In this parameter, enclose the decimal point character in double quotation marks.

DEFAULTCHAR * Single character

If any character field is found by the nonrelational data sourcing routines to be invalid, this character replaces the invalid character.

DEFAULTDATE 19800101 Eight numeric characters

If any component of a date is missing, the missing component from the date is taken from this field.

DISP None N,O,S,MK,D,P,C,UK,D,P,C,U

Default Initial target data set disposition:- N. NEW.- O. OLD.- S. SHR.- M. MOD.

Normal end of job data set disposition:- K. KEEP.- D. DELETE.- P. PASS.- C. CATALOG.- U. UNCATALOG.

Failed job data set disposition:- K. KEEP.- D. DELETE.- P. PASS.- C. CATALOG.- U. UNCATALOG.

MVS only.

DMX_DIR Depends on platform

Maximum is 64 bytes in length.

Specifies the fully qualified path of the disk directory that contains the data maps.The defaults are:- Windows: .\datamaps- Linux and UNIX: ./datamaps- MVS: DD:DATAMAP- i5/OS library: STDATAMAPS

DM_RESOURCE DATASET Default RACF security class of FACILITY uses DTL.DATAMAP.DATASET as default. If this default is not suitable, code:DM_RESOURCE=xxxxx Where xxxxx completes the class name shown by DTL.DATAMAP.xxxxx.MVS only.


Description


DM_SUBTASK N Y or N DM_SUBTASK=Y causes the DATAMAP VSAM file to be opened and owned by a PowerExchange Listener subtask. This option reduces the overhead of file open and close operations. One PowerExchange Listener specifies DM_SUBTASK=Y and works as a data map server. Subsequent PowerExchange Listeners specify DATAMAP_SERVER=<nodename> pointing to the PowerExchange Listener server.

The server reports a change in file status (DTL-00224) and that the PowerExchange Listener is the data map server (DTL-02804).This parameter should be set only on the Server machine, and is not required in any PowerExchange client DBMOVER configuration files.If running with security in DBMOVER parameter SECURITY=(2,x), a profile in the RACF FACILITY class (default DTL.DATAMAP.DATASET) will be required. Users will need access to this class. If the default RACF FACILITY class was not used then specify DM_RESOURCE=xxxxx to specify DTL.DATAMAP.xxxxx, and RACF_CLASS=yyyyy.MVS only.

DTLMSG_CODEPAGE UTF-8 Code page identifier

To use localized Japanese messages, specify DTLMSG_CODEPAGE=CP943.PowerExchange uses the multibyte version of the messages file, dtlmsg_SHIFT-JIS.txt.If you have not specified this parameter, PowerExchange uses the UTF-8 version of the messages file, dtlmsg.txt.If you specify:DTLMSG_CODEPAGE=UTF-8PowerExchange uses the UTF-8 version of the messages file, dtlmsg.txt.

ENCRYPT N N, Y, DES, or RC2

Indicates use of encryption when moving data. Valid values are:- N, or omitting the parameter. No

encryption is used when moving data.- Y. The data is encrypted using a special

Informatica developed algorithm if it is being accessed by way of a PowerExchange Listener.

- DES. The data is encrypted using Digital Encryption Standard (DES) algorithm. If this is selected it must be accompanied by a corresponding ENCRYPTLEVEL parameter.

- RC2. The data is encrypted using RSA Security's algorithm. If this is selected it must be accompanied by a corresponding ENCRYPTLEVEL parameter.


Description


ENCRYPTLEVEL 1, 2, or 3 Must be used if the ENCRYPT parameter is set to DES or RC2.Valid values are:

1. The data is encrypted using:- For DES: 56-bit key encryption- For RC2: 56-bit key encryption

2. The data is encrypted using:- For DES: Triple DES 164-bit key

encryption- For RC2: 64-bit key encryption

3. The data is encrypted using:- For DES: Triple DES 164-bit key

encryption- For RC2: 128-bit key encryption

ENQMAJORNAME DETAIL The Major Name assigned to any PowerExchange task will be the default PowerExchange unless overwritten with this parameter.

ENQSYSTEMS N Y or N By default the ENQMAJORNAME is effective for this LPAR only. If Y is specified this becomes effective across a sysplex.

ERRROWNOTFOUND N Y or N You can specify the Update and Delete behavior for non-existent rows by setting the ERRROWNOTFOUND DBMOVER configuration file parameter on the target platform.For more information about configuring asynchronous write and update and delete behavior, see the PowerExchange Bulk Data Movement Guide.

EXT_CP_SUPPT N Y or N If set to Y, PowerExchange uses its built-in extended character set code pages.

GDGLOCATE N Y or N - Y. Enforces a read of the catalog to pick up the latest GDG information when a file is to be accessed.

- N. Ensures the GDG information remains the same for the duration of the PowerExchange Listener execution.

ICUALIAS See “ICU-specific Configuration Parameters” on page 62.

ICUCNVPROPERTY See “ICU-specific Configuration Parameters” on page 62.

ICUCONVERTER See “ICU-specific Configuration Parameters” on page 62.

ICUDATADIR See “ICU-specific Configuration Parameters” on page 62.


Description


IMSID None This parameter defines the following:- For IMS log-based CDC, defines the IMS

subsystem ID (SSID), the DBDLIB data set, and the RECON data sets. Enables PowerExchange to register IMS data sources for CDC.

- For IMS bulk data movement, defines the IMS SSID and the DBDLIB data set.

Note: This parameter is not used for data map creation.

Syntax:IMSID=(ims_ssid,DBD_lib, RECON=(recon,recon,...))

Where:- ims_ssid. The IMS SSID. Specify a

1- to 8-character alphanumeric string. In the IMS data map, enter an IMS SSID value that matches this value to enable PowerExchange to find the DBDLIB data set.

- DBD_lib. The name of the IMS DBDLIB data set that contains the DBD load modules.

- recon. An IMS RECON data set. This subparameter is required for IMS log-based CDC only. Use commas to separate RECON data sets.

JOBCLASS None Single alphabetic or *

IMS Job submission Class.MVS only.


Description


LISTENER None Maximum 64 character string specifying the node name.

Defines a PowerExchange Listener on the current platform.A maximum of ten LISTENER statements may be coded. For example:LISTENER=(node1,TCPIP, 2480)

Node name Node name of the PowerExchange Listener.

TCPIP Mandatory parameter.

Port number 1 through 65536, port number for the PowerExchange Listener to listen on.

Further parameters are optional:

(a) Send buffer size 1k-1mb

Sets the TCP/IP send buffer size to this value.

(b) Receive buffer size 1k-1mb

Sets the TCP/IP receive buffer size to this value.

(c) Send message size,512 bytes-1mb

Every send message is broken down into units of this size.

(d) Receive message size, 512 bytes -1mb

Every receive message is read in units of this size.

(e) Long receive time, 1-14400 in seconds

If PowerExchange determines that a long wait is required, the receive timeout value is set to this value for this request only.

(f) IP address string.

(g) SSL SSL parameter. For more information about the SSL parameter, see “Secure Sockets Layer Support” on page 105.

LOADCTLFILE None Data set name PDS holding the control card template for DB2 for z/OS bulk data loads.

LOADJOBFILE None Data set name PDS holding the JCL template for DB2 for z/OS bulk data loads or the IDMS metadata skeleton job IDMSMJCL.

LOG_CODEPAGE UTF-8 Code page identifier

Windows programs can read log records correctly if they are all in the same code page. If you want messages written to the PowerExchange log file to be in a different code page to the control code page, then you must specify LOG_CODEPAGE.For example:On Windows, if you want the PowerExchange log file in Japanese, specify LOG_CODEPAGE=CP943.

LOGNODE None Maximum is 64 bytes long

Unused.


Description


LOGPATH None Valid path Path to PowerExchange message log files on Linux, UNIX, and Windows systems.If this parameter is not specified, PowerExchange writes log files in the current working directory. You can use the DETAIL_LOGPATH environment variable to override this path.

LOGSID Specifies the location of the IDMS logs and the PowerExchange log catalog. This statement is required in the DBMOVER configuration file on the platform on which the catalog resides.Syntax:LOGSID=( registration logsid, listener node, log catalog name, log catalog instance name)

Where:- registration logsid. The logsid value

given on the Add Registration Group screen when registering for capture.

- listener node. The node upon which the request will be made. This will be the node specified in the Listener= statement at the top of the DBMOVER configuration file.

- log catalog name. The data set name of the PowerExchange Log Catalog.

- log catalog instance name. The name given to the log entry when running utility DTLULCAT which creates the input for DTLULOGC.

LRECL None 4-32756 If a new file is dynamically allocated and no LRECL is supplied this value will be used.MVS only.

MAXTASKS 5 1-255 Maximum number of concurrent tasks that a PowerExchange Listener will start. The actual number of subtasks supported may be constrained by available memory or other operating system resource constraints and may be substantially less than 255.

MQAUTOCREATE N Y or N If MQAUTOCREATE = Y is set in the configuration file when a data movement command is run:- If the application specified is not found in

the restart queue, it will be automatically added.

- If a source-target combination is not found, one will be automatically added.

MQRECOVER Valid values are from 0 to 86400.

Number of seconds after which an application marked as “running” is considered to have terminated abnormally.Specifying zero means that an application marked as ‘running’ will never be considered to have abnormally terminated.


Description


MQRESTART queue_name Specifies the MQSERIES Restart Queue to which restart tokens will be written. This is used when using MQSERIES as the target of the change data capture.

MSGPREFIX PWX Alphanumeric character string

Specifies the message prefix string. Maximum is eight characters.

MSGPREFIX-HYPHEN Y Y or N Specifies whether or not PowerExchange outputs a hyphen character between the message prefix and message number.For example:- MSGPREFIX-HYPHEN=Y results in

PWX-06111.- MSGPREFIX-HYPHEN=N results in

PWX06111.

MVSDB2AF CAF CAF or RRSAF

On MVS, the type of DB2 attachment facility that PowerExchange uses to connect to DB2 for z/OS.

If you specify RRSAF,RRS must be configured and running on the MVS system.

For more information, see “PowerExchange Security” on page 85.

NEGSIGN - Single character

Any negative numeric fields will have this character as the - sign.


Description


NETPORTSee notes on IMS data sourcing.Required if tape data to be accessed. Not required if the access method is IMS ODBA.

None Associates a Netport job with a port. You can use a Netport job in special situations, for example, to access sources through a non-PowerExchange module such as IMS DLIBATCH, alleviate excessive wait times for tape mounts, perform GDG processing, or provide IDMS access with full user ID checking. Syntax example:NETPORT=(node1,9999,,,"INFA.Vxyz.RUNLIB(IMSJCL)",mypsb,node1,9999,ims)

Where the positional parameters are:- First parameter. Node name matching

LISTENER. This node name must match the node name in the PARM found in the EXEC DTLLST statement within the PowerExchange Listener JCL otherwise the PowerExchange Listener will try and satisfy the request as a non-Netport task.

- Second parameter. Port number that matches the PowerExchange Listener port number.

- Third parameter (optional). Node from which to take further TCP/IP characteristics. Defaults to parameter 1.

- Fourth parameter (optional).Port number from which to take further TCP/IP characteristics. Defaults to parameter 2.

- Fifth parameter. Name of the JCL member to be submitted.

- Additional parameters 6 through 13 (up to eight). Further substitutions.

MVS only.


Description


NODE None Maximum 64-character string

Specifies the details of remote PowerExchange Listeners that you can connect to.For example:NODE=(MVS1,TCPIP,MVS1, 2480)

Node name Node name of the PowerExchange Listener.

TCPIP Mandatory parameter.

IP Address The host name or IP address of the PowerExchange Listener to be contacted.

Port number The port number that the PowerExchange Listener listens on.1 through 65536.

Further parameters are optional:

(a) Send buffer size 1k-1mb

Sets the TCP/IP send buffer size to this value.

(b) Receive buffer size 1k-1mb

Sets the TCP/IP receive buffer size to this value.

(c) Send message size,512 bytes-1mb

Every send message is broken down into units of this size.

(d) Receive message size, 512 bytes -1mb

Every receive message is read in units of this size.

(e) Long receive time, 1 through 14400 in seconds

If PowerExchange sees that a “long wait” is needed the receive timeout value is set to this value for this request only.

(g) SSL or ZOSSSL

SSL parameter. For more information, see “Secure Sockets Layer Support” on page 105.

NOGETHOSTBYNAME N Y or N When Y is specified, PowerExchange does not attempt to obtain the IP address of the local machine. PowerExchange requires this for license checking and diagnostic messages. However in certain situations, normally because of a failure to add an entry in the TCP/IP HOSTS table for the local machine, resolution of the IP address can take a long time, resulting in poor performance. Therefore, you can temporarily suppress this behavior until the problem can be resolved. This parameter should only be set under guidance from Informatica Global Customer Support as it also requires use of a special license key.


Description


NRDB_WRITE_CHAR_NULL_FILL

Space Character value or hex value

For NRDB or NRDB2, specify as NRDB_WRITE_CHAR_NULL_FILL= A (or Hex C1) as an example. This enables null character fields to be delivered with a filler of choice. To enable this to be used the field in the data map must be defined as a “nullable field”. Low values can be specified in the hex format.

NRDB_WRITE_NUM_NULL_FILL

0 Numeric value or hex value

For NRDB or NRDB2, specify as NRDB_WRITE_NUM_NULL_FILL= 9 (or Hex F9) as an example. This enables null character fields to be delivered with a filler of choice. To enable this to be used the field in the data map must be defined as a ‘nullable field’. Low values can be specified in the hex format.This is valid for unpacked numeric formats.

NUMERICSIGN C C or F PowerExchange will, by default, deliver unsigned fields as positive X'C'. When working with EBCDIC data, coding NUMERICSIGN=F in the DBMOVER configuration file of the client, regardless of platform, will force unsigned fields to be returned unsigned X'F'.

ODBASUPP None YES Set to 'YES' to enable IMS ODBA.

OFFLOADPROCESSING D - Y- A- N- D

Offload processing can be used only if the data: - Is being read only.- Has the type of VSAM (when processed

sequentially), SEQ, or TAPE.This parameter in the configuration file can be overridden by the Offload parameter in the data map itself.

Valid values are:- Y. All data map processing are moved to

the client CPU. With data maps that have substantial processing requirements (i.e. variable fields) significant CPU savings will be gained on the server with a slight increase in network usage.

- A. PowerExchange decides whether to offload the data map processing. Offload processing will be used if the source contains variable length fields.

- N or D, or the parameter is omitted. The move does not occur.

Note: Offload processing will not take place if the data map is being used for Parameterized SQL or if Data Checking has been selected for the map.

ORA_ERRORFILE Character string composed of path and file_name

Specifies the path and filename of the Fault Tolerance error files.For more information about ORA_ERRORFILE, see the PowerExchange Bulk Data Movement Guide.


Description


ORACLECODEPAGE Defines the PowerExchange code page and PowerCenter code page that are used for a particular Oracle database. ORACLECODEPAGE=(<tnsname>,<pwxcp>,pccp) For example:ORACLECODEPAGE=(BU920DTL, UTF-8,UTF-8)The code pages that are specified must match the environment variable NLS_LANG.You can specify up to 20 ORACLECODEPAGE parameters for separate databases/TNSNAMES.

ORACLEDB None, N, N Character string, {Y|N}, {Y|N}

Specifies the Oracle database name that is used for data capture. Syntax:ORACLEDB=(database name, {Y|N}, {Y|N})

Where:- database name. Name of the database

that holds the data to be captured. This is the name of the database, not the name of the instance. This parameter is mandatory and no wild cards are allowed.

- {Y|N}. Optional. Specifies the format of the Oracle Capture restart token. Database names are stored as part of the restart token. <N> (optional). A placeholder and must be N if the third sub-parameter is specified. Otherwise, it is optional. <Y>. Specifies that the database name is stored in the format specified in the client code page.

- {Y|N}. Optional. Indicates how database name is stored. <N>. Specifies that the database name is stored as a UTF-8 string. Only use this parameter when specifically directed to do so by Informatica Global Customer Support.


Description


ORACLEIDSee the PowerExchange CDC Guide for Linux, UNIX, and Windows for details.

Specifies the Oracle instance name and connection information. PowerExchange requires an ORACLEID statement for each Oracle instance for which you capture and extract change data. Specify this statement in the dbmover.cfg file that is on the machine where your PowerExchange extractions or PowerExchange Logger processes run. You can specify a maximum of 20 ORACLEID statements in a single dbmover.cfg file. Syntax:ORACLEID=(collection_id,Oracle_sid,connect_string, capture_connect_string

Where:capture_connect_string is a database service name, such as OEMREP.INTERNAL.INFA.COM, that PowerExchange uses to connect to the Oracle_sid that is specified in the same ORACLEID statement. This parameter is required in the following situations:- When the PowerExchange extraction session or PowerExchange Logger

process that extracts Oracle change data runs on a machine that is remote from the Oracle instance.

- When the PowerExchange extraction session or PowerExchange Logger process that extracts Oracle change data runs on a machine that has more than one Oracle instance and the instance that contains the data to be captured is not the default instance on that machine. The default instance is specified by the ORACLE_SID environment variable.

Specify this parameter in the ORACLEID statement of the dbmover.cfg file that is on the machine where the Oracle data is extracted.

collection_id is an Oracle instance identifier that matches the collection ID that is specified in a capture registration for an Oracle source table.

connect_string is a database service name, such as OEMREP.INTERNAL.INFA.COM, that enables the PowerExchange Navigator to connect to the Oracle_sid that is specified in the same ORACLEID statement. This parameter is required in the following situations, after you have created capture registrations: - When the PowerExchange Listener to be used for retrieving Oracle data is

remote from the Oracle instance.- When the PowerExchange Listener to be used for retrieving Oracle data exists

on a machine that has more than one Oracle instance, and the instance that contains the data to be captured is not the default instance on that machine. The default instance is specified by the ORACLE_SID environment variable.

Specify this parameter in the dbmover.cfg file of the PowerExchange Listener involved in retrieving the Oracle data. If you plan to run a row test on the capture registrations, also specify the capture_connect_string parameter in the same dbmover.cfg file.

Oracle_sid is name of the Oracle database that contains the tables that you registered for change data capture.

OUSP N Y or N Set to Y if using ADABAS and PowerExchange SECURITY=(2,Y). Set this parameter only if instructed to do so by Informatica Global Customer Support.


Description


PC_AUTH N Y or N An IDMS specific parameter. Code PC_AUTH=Y if APF authorized copies of IDMS load libraries are NOT to be used. This will invoke program control authorization. For more information about the PowerExchange Listener, see the PowerExchange Bulk Data Movement Guide.Note: If using Netport jobs to access data then the Netport job will require a DTLLOAD DD statement which will point to the IDMS load libraries if PC_AUTH is set to Y. For more information about Netport jobs, see “Netport Jobs” on page 67.MVS only.

PIPE | Single character

If a TXT output file is generated, this character will separate every field.

POLLTIME 2000 250-15000 Time in thousandths of a second that the PowerExchange Listener will wait between polling for any outstanding connections. 1000 = one second.

PRE861_COMPAT Y Y or N Controls whether PowerExchange automatically attempts to use offload processing.

Valid values are:- N. If you select Auto for the Offload

Processing attribute on PWXPC connections, PowerExchange uses offload processing if possible. In addition, PowerExchange ignores the Offload Processing option specified in VSAM and sequential data maps for previous releases.

- Y. PowerExchange uses offload processing only if you select it. On PWX DB2zOS relational connections or PWX CDC Real Time application connections, select Yes for the Offload Processing attribute. On PWX NRDB Batch application connections, select either Filter Before or Filter After for the Offload Processing attribute.

Tip: You must perform additional tasks to use offload processing for bulk data movement operations or CDC. To control when PowerExchange uses offload processing, set PRE861_COMPAT to Y, or allow it to default to Y. For more information, see:- PowerExchange Bulk Data Movement

Guide - The PowerExchange CDC guide for your

operating system

PRGIND N Y or N Enables progress, such as rows read, to be viewed in the command line output messages. If Y is specified progress indication will be displayed as the request is being run.If N is specified no progress is displayed.


Description


PRGINT 250 Numeric Used in conjunction with PRGIND=Y. Specifies the interval (in rows read) at which the progress message is displayed. For example:PRGIND=Y PRGINT=100 Displays the progress message after every 100 rows have been read.If no PRGINT is specified then the default (250) is used.

PWXSOMAXCONN Operating system SOMAXCONN value

Positive integer

Specifies the maximum number of TCP/IP connections for sockets that PowerExchange uses to listen for connections.

RACF_CLASS FACILITY Format RACF_CLASS=yyyyy, where yyyyy overrides the default FACILITY class for RACF definitions.MVS only.

RDBMSINSRTDFLT N Y or N Determines whether to use PowerExchange default values or DB2 default values for DB2 columns that are defined with the WITH DEFAULT clause.Valid values are:- Y. Use the DB2 default values. You must

have defined the columns with a clause that enables DB2 to supply a default. Otherwise, an SQL error is generated.

- N. Use the PowerExchange default values.

RECFM None F, V, FU, FB, VU, VB, FBA, VBA

Used if a new file is dynamically allocated and no RECFM is supplied.MVS only.

REJECT_FILE_DELIMITER Comma (,) Char The default delimiter in a reject file (Asynchronous Write with Fault Tolerance) is a comma (,).This parameter overrides the default delimiter. This is useful if the data being written contains commas.REJECT_FILE_DELIMITER=?If a delimiter of semicolon (;) is required, then this has to be surrounded by double quotes in the parameter. For example: REJECT_FILE_DELIMITER=”;”For more information about reject files, see the PowerExchange Bulk Data Movement Guide.

RELEASE None Y Sets default to free unused space for new file allocation.MVS only.


Description


RMTRDBDIRE Value set by CRTPWXENV command at installation. Default is *LOCAL.

Maximum length of 18 characters.

In a remote journaling environment, the name of the DB2 database on the local system that contains the DB2 source tables for CDC. This value must be defined to i5/OS with the Work with Relational Database Directory Entries (WRKRDBDIRE) function.

Include this parameter only if you plan to use remote journal receivers for CDC.

When you run the CRTPWXENV command during installation, this parameter is automatically populated with either the the default value of *LOCAL or the value that you specify in the optional RMTRDBDIRE parameter on the CRTPWXENV command. You can use this DBMOVER parameter to override the original value, if necessary.

RMTSYSNAME Value set by CRTPWXENV command at installation. Default is *NONE.

Maximum length of 68 characters.

In a remote journaling environment, the name of the i5/OS host that contains the DB2 source tables for CDC and the local journals and journal receivers. Also called the local system. This value must be defined to i5/OS with the “Work with TCP/IP Host Table Entries” function.

Include this parameter only if you plan to use remote journal receivers for CDC.

When you run the CRTPWXENV command during installation, this parameter is automatically populated with either the the default value of *LOCAL or the value that you specify in the optional RMTRDBDIRE parameter on the CRTPWXENV command. You can use this DBMOVER parameter to override the original value, if necessary.


Description


SECURITY 0,N Number range 0-2, N or Y

Used only on MVS and i5/OS to define which security checks will be performed.The first parameter controls the use of authentication and MVS or i5/OS facilities for resource control.The second parameter controls the use of the PowerExchange selective sign-on facility.

Syntax:SECURITY=(n,{Y|N})First parameter:- 0. Does not validate the user ID and

password that are passed from the attached system. All resource access is carried out using the user ID under which the PowerExchange Listener is running.

- 1. Validates that the user ID and password passed from the attached system are known to the operating system. However, all resource access is carried out using the user ID under which the PowerExchange Listener is running. Requires the STEPLIB to be APF-authorized for the PowerExchange Listener and any NETPORT jobs.

- 2. Validates that the user ID and password passed from the attached system are known to the operating system. All resource access is carried out using this user ID. Requires the STEPLIB to be APF-authorized for the PowerExchange Listener and any NETPORT jobs.

Second parameter:- Y. Invokes the PowerExchange selective

signon facility to verify that the user ID passed from the attached system is allowed to access the PowerExchange Listener. For more information, see “PowerExchange Security” on page 85.

- N. Does not invoke the selective signon processor.

SHOW_THREAD_PERF 10000 through 50000000

Number of change records PowerExchange processes in a statistics reporting interval. After processing the change records, PowerExchange writes messages PWX-31524 through PWX-31259 to the PowerExchange log file. If you select the Retrieve PWX log entries attribute on a PWX CDC application connection, PWXPC writes the messages in the session log. The messages include information that can be used to tune multithreaded processing.


Description


SPACE None Space type:- C.Cylinders.- T.Tracks.Space allocation:- 1 to 512

primary- 1 to 512

secondary

Specifies the default that is used if a new file is dynamically allocated and no space allocation is supplied, this value is used. MVS only.

SSL None PASS, KEY, CALIST, CAPATH

Specifies the SSL certificate that you are using to make the SSL connection.

SSL_ALLOW_SELFSIGNED N Y or N SSL_ALLOW_SELFSIGNED=Y enables the use of self-signed certificates which are locally generated certificates for testing and demonstration purposes. Self-signed certificates are not verifiable by following the chain to a trusted certificate authority.By default, it is necessary to use certificates from a commercial certificate authority trusted by the system.

SSL_REQ_CLNT_CERT Y Y or N SSL_REQ_CLNT_CERT=Y requests the certificate from the Client.

SSL_REQ_SRVR_CERT N Y or N SSL_REQ_SRVR_CERT=Y requests the certificate from the Server.

STATS None SMF or FILE Determines if statistics records are to be written to SMF or to a file.Specified as:STATS=(SMF,<record number>, <interval>)orSTATS=(FILE,<filename>, <interval>).For more information see “Statistics Logging Including SMF” on page 77.

SUBMITTIMEOUT 60 1-86400 The number of seconds that the PowerExchange Listener will wait before automatically timing out a submitted batch job that has not started. This parameter applies to all batch jobs spawned by the PowerExchange Listener which includes Netport, DB2 bulk load (DB2LDJCL), IDMS metadata (IDMSMJCL), and DTLREXE PROG=SUBMIT jobs. The following message will be issued in the PowerExchange Listener and to the client if the job doesn’t start in the specified period: PWX-00426 Failed to start task - Job timed out.

Note: Ensure that this value is less than the send/receive TIMEOUTS= values. Otherwise, the TIMEOUTS value will occur first and message PWX-00426 will not be issued.MVS only.


Description


SUPPRESS_DATA_ LOGGING

None Y The valid value is Y.Displays “Data print suppressed” instead of the error data that is normally displayed in the message.

SVCNODE On Linux, UNIX, and Windows, use the SVCNODE statement to specify the port on which a PowerExchange service listens for pwxcmd commands.

For more information, see “pwxcmd Command Configuration Statements” on page 64.

SYSOUT_TIMESTAMP N Y or N Prefixes all sysout output with the date timestamp. This is the same as is written to the PWX log except that no platform information is output. Format: ddmmyy hhmmss PWXnnnn <msg_text>

TAPEWAIT N Y or N If specified as TAPEWAIT=Y, then all jobs (except server jobs such as the PowerExchange Listener) will wait for tape mounts, if required, issuing an operator message giving the option to cancel wait. This parameter can be used in all environments but cannot be used in any subtasks running under the PowerExchange Listener. However it can be used for separate jobs (such as Netport jobs) submitted by the PowerExchange Listener and running in their own address space. Therefore any subtask that requires this parameter could be run as a Netport jobNote: Can only be used if the library is APF authorized.MVS only.

TCPIP_ASYNC N Y or N Defines whether PowerExchange uses asynchronous network I/O when reading change data. If you specify Y, PowerExchange writes change data to the network buffers and reads change data from the change stream asynchronously. Set this parameter to Y in the DBMOVER configuration file on the source system to help improve throughput for CDC sessions.This parameter is not supported on AIX, i5/OS, or Windows.

TCPIPBUFSIZE 128000 34816- 1048576

Defines the maximum send and receive TCP/IP buffer size. You can override the maximum buffer size by entering different values for the NODE in the PowerExchange Listener parameters.

TCPIPVER None 2 Set this parameter to 2 only if you use CA TCPAccess rather than IBM MVS TCP/IP. If you omit this parameter, the standard MVS TCP/IP stack is used.


Description


TEMPHLQ None Up to 17 characters

Defines the High Level Qualifier for temporary data sets created for the download of IDMS metadata when SECURITY is 0 or 1 (ignored where SECURITY=2). This is used to override the PowerExchange Listener user id for sites that do not want data sets with IDMS metadata created with the PowerExchange Listener user id as the HLQ. Example:TEMPHLQ=B.CThis creates temporary data set B.C.METADATA.DTL000001 when running an IDMS copybook metadata import.MVS only.

TEXT_EOF_FOR_BINARY N Y or N Normally, for binary files, PowerExchange ignores any hex values such as, x'1a'.To change this behavior and allow the hex value x'1a' to be processed as an EOF indicator, you can set this parameter to TEXT_EOF_FOR_BINARY=Y.

TIMEOUTS 30,30,30 Numeric in the range:

0 to 65535 for connect and send.

-1 to 65535 for receive.

Connect, send and receive timeout. If this time interval (in seconds) is exceeded, PowerExchange cancels the request and indicates a timeout error. The third (receive) subparameter can be set to -1. This value means that no timeout is set for receive. For example: 30,30,-1

TRACE None TRACE will severely impact performance and should only be at the direction of Informatica Global Customer Support.

TRACING Requests that PowerExchange use alternative log files for messages.

For more information, see “PowerExchange Message Log Options” on page 36 and “TRACING Statement” on page 37.

UNIT None 6-character string

Default Unit value. If a new file is dynamically allocated, this value is used for the unit.

For IDMS, this must be set to a value which will allow intermediate metadata MVS data sets to be allocated.MVS only.

VOLSER None 6-character string

Default VOLSER that is used when a new file is dynamically allocated. For IDMS, this VOLSER must be a value that enables the allocation of intermediate metadata data sets.MVS only.


Description


VSAM BUFNI=1,BUFND=2

VSAM=(BUFNI=<1 to 255>, BUFND=<0 and 2 to 255>)BUFNISpecifies the number of I/O buffers VSAM is to use for transmitting the contents of index entries between virtual and auxiliary storage for keyed access. A buffer is the size of a control interval in the index. BUFNI must be either:- A number between 1 and 255. Number

of buffers.- 0. The operating system default is used.

The default is the minimum number required. Additional index buffers improve performance by providing for the residency of some or all of the high-level index, thereby minimizing the number of high-level index records retrieved from DASD for key-direct processing. For more information on optimizing performance, see IBM's OS/390 DFSMS: Using Data Sets. The default is the minimum number required. The maximum number of buffers allowed is currently 255 (254 data buffers and 1 insert buffer).

BUFNDSpecifies the number of I/O buffers VSAM is to use for transmitting data between virtual and auxiliary storage. A buffer is the size of a control interval in the data component. BUFND must be one of the following values:- A number between 2 and 255. Number

of buffers.- 0. The operating system default is used.- 1. Not allowed.

The default is the minimum number required. Note, however, that minimum buffer specification does not provide optimum sequential processing performance. Generally, the more data buffers specified, the better the performance.Note also that additional data buffers benefits direct inserts or updates during control area splits and benefits spanned record accessing. The maximum number of buffers allowed is currently 255 (254 data buffers and 1 insert buffer). See IBM's OS/390 DFSMS: Using Data Sets for more information on optimizing performance and system-managed buffering.MVS only.


Description


PowerExchange Message Log Options

PowerExchange uses log files to store informational and error messages about bulk and change data capture (CDC) operations. You can configure PowerExchange to use the following log files for these messages:

♦ PowerExchange log file. The PowerExchange log file is the default log file for all PowerExchange messages. PowerExchange creates this log file, if it does not exist. If the PowerExchange log file becomes full, you must clear it. PowerExchange writes each message, as it occurs, to this single log file.

♦ PowerExchange alternative log files. You can request that PowerExchange use alternative log files by coding the TRACING statement in the DBMOVER configuration file. PowerExchange creates these alternative log files, if they do not exist. PowerExchange rotates through the alternative log files, which eliminates out-of-space conditions. Also, PowerExchange buffers messages and writes them to the disk files based on a customized flush interval, thereby reducing I/O activity for these files.

PowerExchange Log FileBy default, PowerExchange writes all messages to the PowerExchange log file. PowerExchange creates this file if it does not exist. The following table shows the name and default location of the PowerExchange log file by platform:

Writing messages to a single file means that, without manual intervention, the volume of data continues to grow until the PowerExchange log file becomes full. This out-of-space issue usually occurs for long-running PowerExchange tasks, such as the PowerExchange Listener. Also, PowerExchange must open and close the PowerExchange log file to write each message. This open and close activity can decrease performance and increase resource utilization when message volume is high.

WAITDSN N Y or N WAITDSN=Y causes the job to wait for the in-use data set to be released.This parameter can be used in all environments but cannot be used in any subtasks running under the PowerExchange Listener. However it can be used for separate jobs (such as Netport jobs) submitted by the PowerExchange Listener and running in their own address space. Therefore any subtask that requires this parameter could be run as a Netport jobNote: Can only be used if the library is APF authorized.MVS only.

Platform File Name Location

i5/OS datalib/LOG(DTLLOG) A member of the LOG file in the PowerExchange data library, datalib, which PowerExchange creates during installation.

Linux, UNIX, and Windows

detail.log A file in the current working directory of the PowerExchange process that is running.

MVS DTLLOG DD statement A data set that is allocated to the DTLLOG DD statement in the PowerExchange job. This DD statement can point to a permanent data set or to a SYSOUT data set.


Description


PowerExchange Alternative Log FilesPowerExchange alternative log files provide more flexibility and better performance than the original PowerExchange log file. If you configure PowerExchange to use alternative log files, PowerExchange writes most messages to these log files. Only the initial startup messages are still written to the original PowerExchange log file. After PowerExchange initializes the tracing subtask, it only writes messages to the alternative log files.

PowerExchange alternative log files provide the following benefits over the original PowerExchange log:

♦ PowerExchange can optionally use multiple files for logging messages. When the current log file is full, PowerExchange switches to the next log file, thereby preventing log-full conditions.

♦ After opening an alternative log file, PowerExchange buffers messages so that they can be written to the log file at specified intervals. This buffering reduces the amount of open, close, and write activity on the file to improve performance and resource use.

♦ You can control the size and number of alternative log files and the frequency at which the log records are flushed to the file.

To use alternative log files, specify the TRACING statement in the DBMOVER configuration file.

TRACING Statement

The TRACING statement has the following syntax:

TRACING=( PFX=prefix, [APPEND=Y|N], [BUFFERS=number_of_buffers], [FILENUM=number_of_files], [FLUSH=flush_interval], [RECLEN=record_length], [SIZE=log_size], [VIEW=Y|N])

The TRACING statement has the following parameters:

PFX=prefixRequired. Specifies the prefix for the alternative log file names.

PowerExchange uses the following platform-based rules to create the alternative log file names:

i5/OS PowerExchange uses the PFX value to create the member names of the log files in the PowerExchange data library. The log file names vary based on whether the PowerExchange Listener, PowerExchange Condense, or other PowerExchange jobs create the files.

♦ The PowerExchange Listener uses the following file naming convention:

datalib/Plistener_port(prefixnnn)

♦ PowerExchange Condense and other PowerExchange jobs use the following file naming convention:

datalib/JOBjob_num(prefixnnn)

These naming conventions include the following variables:

♦ datalib. The PowerExchange data library name specified during PowerExchange installation.

♦ job_num. The i5/OS job number for the tracing subtask, DTLTRTSK, that runs under PowerExchange Condense or other PowerExchange jobs.

♦ listener_port. The PowerExchange Listener port number.

♦ nnn. A sequential number from 001 through 999.

PowerExchange Message Log Options 37

For example, a PowerExchange Listener that has a listener port number of 2480, a PFX value of PWXLOG, and a FILENUM value of 3 creates the following log files:

datalib/P02480(PWXLOG001)datalib/P02480(PWXLOG002)datalib/P02480(PWXLOG003)

Maximum length for the PFX value is 7 characters.

MVSPowerExchange uses the PFX value as the high-level qualifier (HLQ) for the sequential log data sets. By default, PowerExchange uses dynamic allocation to create the alternative log data sets. You can also use JCL DD statements to create and allocate alternative log data sets. For more information about allocation options for MVS log data sets, see “Allocation Options for Alternative Log Data Sets on MVS” on page 41.

You cannot use DFSMS compression for alternative log data sets.

The log data set names vary based on whether the PowerExchange Listener or other PowerExchange batch jobs or started tasks create the files.

♦ The PowerExchange Listener uses the following file naming convention:

prefix.sysid.Plistener_port.Nnnn

♦ All other PowerExchange batch jobs and started tasks use the following file naming convention:

prefix.job_name.job_num.sysid.Nnnn


♦ job_name. The job name of the batch job or started task.

♦ job_num. The JES job number, which begins with JOB for batch jobs and STC for started tasks.



♦ sysid. The MVS system ID of the system on which the batch job or started task runs.

For example, a PowerExchange Listener that runs on system MVS1 with a port number of 2480, a PFX value of PWXLOG, and a FILENUM value of 3 creates the following log files:

PWXLOG.MVS1.P02480.N001PWXLOG.MVS1.P02480.N002PWXLOG.MVS1.P02480.N003


Linux, UNIX, and Windows PowerExchange uses the PFX value as the subdirectory name in which to place the log files. PowerExchange uses the LOGPATH parameter in the dbmover.cfg file to determine the directory in which to place this log subdirectory. For more information about the LOGPATH parameter, see “DBMOVER Parameter Descriptions” on page 4.

The log file names vary based on whether the PowerExchange Listener, PowerExchange Condense, or other PowerExchange tasks create the files.

♦ The PowerExchange Listener uses the following file-naming convention:

Linux and UNIX: logpath/prefix/DTLLST1.plistener_port.nnnn.log

Windows Listener Service: logpath\prefix\DTLLSTNT.plistener_port.nnnn.log

Windows Listener: logpath\prefix\DTLLST1.plistener_port.nnnn.log

♦ The PowerExchange Logger for Linux, UNIX, and Windows uses the following file-naming convention:

Linux and UNIX: logpath/prefix/DTLCACON.tyyyymmddhhmmss.ppid.nnnn.log

Windows: logpath\prefix\DTLCACON.tyyyymmddhhmmss.ppid.nnnn.log


♦ For other tasks, PowerExchange uses the following file-naming convention:

Linux and UNIX: logpath/prefix/module.tyyyymmddhhmmss.ppid.nnnn.log

Windows: llogpath\prefix\module.tyyyymmddhhmmss.ppid.nnnn.log


♦ logpath. The value of the LOGPATH parameter in the dbmover.cfg file.


♦ module. The name of the PowerExchange module that is running, such as DTLURDMO for that utility or DTLODBCDRVR for PowerCenter operations.


♦ pid. The process ID of the PowerExchange task.

♦ yyyymmddhhmmss. The timestamp when the file is created.

For example, a PowerExchange Listener that runs on UNIX with a port number of 2480, a PFX value of PWXLOG, and a FILENUM value of 3 creates the following log files:

logpath/PWXLOG/DTLLST1.p02480.n001.loglogpath/PWXLOG/DTLLST1.p02480.n002.loglogpath/PWXLOG/DTLLST1.p02480.n003.log


APPEND={Y | N}Optional. Controls how PowerExchange uses log files when it restarts.

♦ Specify Y to have PowerExchange append the new log records to the log file that was most recently used. If no log files exist, PowerExchange opens a new log file.

♦ Specify N to have PowerExchange overwrite the log file that was least recently used.

Note: On z/OS, the APPEND=Y option is not supported for Generation Data Groups (GDGs). When the PowerExchange Listener starts, it always writes to a new GDG.

Default is Y.

BUFFERS=number_of_buffersOptional. Specifies the number of buffers that PowerExchange allocates to receive message and trace information from PowerExchange subtasks. If the buffer space is full, the PowerExchange subtasks that generate message and trace information wait until buffer space is available. This buffer space is used internally by PowerExchange programs only.

Do not specify this parameter unless instructed to do so by Informatica Global Customer Support.

Valid values are 5 through 9999.

Default is 100.

FILENUM=number_of_filesOptional. Specifies the number of alternative log files that PowerExchange creates and uses. When a log file is full, PowerExchange switches to the oldest alternative log file and overwrites it.

You can specify a number from 1 to 99 for the number of files.

Default is 5.

FLUSH=flush_intervalOptional. Specifies the number of log records that PowerExchange collects before it flushes them to the log file on disk. PowerExchange must periodically flush log records to enable PowerExchange to recover from out-of-space conditions. Low flush values result in more I/O activity to the log file.

Valid values are 1through 99.

Default is 99.


RECLEN=record_lengthOptional. Specifies the record length that PowerExchange uses to write log records to the log file. PowerExchange writes the log record on multiple lines if the length of the message exceeds the record length.


Default is 80.

SIZE=log_sizeOptional. Specifies the approximate amount of log data, in megabytes, that PowerExchange writes to an alternative log file. After PowerExchange reaches this value, it closes the current log file and opens the next log file to continue writing log records.

To determine the number of blocks PowerExchange uses for dynamically allocated log files on MVS, see “Determining the Size of the Alternative Log Data Sets on MVS” on page 40.

Note: On MVS, PowerExchange limits the amount of log data it writes to manually allocated log data sets to the SIZE value when the data set exceeds the SIZE value. If the data set is smaller than the SIZE value, the data set size limits the amount of log data PowerExchange writes.


Default is 100.

VIEW={Y|N}Optional. Controls whether PowerExchange periodically closes and reopens the current log file. You can specify this parameter on all platforms, but it is most useful on MVS. On MVS, you cannot see any log records in the log data set until it is closed. On other platforms, you can view the log records after PowerExchange flushes them to disk.

On MVS, specify VIEW=Y to periodically close and then reopen the log data set, based on the FLUSH interval.

To use VIEW=Y with GDG log data sets, you must also perform the following tasks:

♦ Specify a relative generation value of 0 in the DTLLOGnn DD statement that references the GDG in the PowerExchange job.

♦ Specify DISP=SHR on the DTLLOGnn DD statement.

♦ Create at least one generation of the GDG data set prior to running the PowerExchange job.

If you have vendor products on MVS that manipulate data set allocations, these products might interfere with VIEW=Y operation. Most commonly, these vendor products change the SYSDSN ENQ to EXCLUSIVE mode, which prevents you from viewing the data set.

Warning: VIEW=Y can adversely impact the performance of the PowerExchange job that writes to the log data set because of frequent data set open and close operations. Use the default value of 99 for the FLUSH parameter to minimize this performance impact.

Default is N.

Determining the Size of the Alternative Log Data Sets on MVS

PowerExchange uses the following DCB attributes to allocate alternative log data sets on MVS:

♦ Block size (BLKSIZE) is 27,998.

♦ Data set organization (DSORG) is PS.

♦ Record format (RECFM) is VB.

♦ Record length (LRECL) is the value of the RECLEN parameter on the TRACING statement.


You can use the following formulas to calculate the primary and secondary space, in blocks, of an alternative log data set:

primary space = log_size / 27998 secondary space = primary_space / 10

These statements include the following variables:

♦ log_size. The value that you specify for the SIZE parameter in the TRACING statement, which has been converted from MB to bytes.

♦ 27998. The block size in bytes.

♦ primary_space. The value calculated by dividing log_size by 27,998.

For example, if you specify 100 for the SIZE parameter, then:

primary space = 104,857,600 / 27,998 = 3745 blockssecondary space = 3745 / 10 = 374 blocks

Allocation Options for Alternative Log Data Sets on MVS

If you specified the TRACING statement in the DBMOVER configuration file for a PowerExchange job or started task, PowerExchange uses alternative log data sets. PowerExchange dynamically allocates the log data sets unless you indicate otherwise.

Use one of the following methods to allocate the alternative log data sets:

♦ Dynamically allocate data sets. PowerExchange dynamically allocates log file data sets unless you define DTLLOGnn DD statements in the JCL for the PowerExchange batch job or started task. To name the dynamically allocated log files, PowerExchange uses the PFX parameter value that you defined in the TRACING statement. For more information about the PFX parameter, see “TRACING Statement” on page 37.

You can control the number of dynamically allocated log files that PowerExchange creates using the FILENUM parameter of the TRACING statement.

♦ Define DD statements with different data set names. You can allocate the log data sets by specifying DTLLOGnn DD statements in the JCL for the PowerExchange batch job or started task. The variable nn is a number from 01 to 99. You can specify up to 99 DTLLOGnn DD statements in the JCL.

If you use DD statements, you must still specify a TRACING statement with the PFX parameter. However, PowerExchange ignores the PFX and the FILENUM values for all jobs that have DTLLOGnn DD statements. PowerExchange uses the same round-robin algorithm for the JCL-specified log data sets as it does with dynamically allocated ones.

♦ Define a DD statement with a GDG data set name. You can use GDG data sets for log data sets. To use GDG data sets, allocate a single DTLLOGnn DD that specifies a GDG based name with a relative generation number.

PowerExchange does not overwrite or rotate through the GDG data sets when a log data set becomes full. Instead, PowerExchange dynamically allocates the next generation of the GDG when it reaches the SIZE value in the TRACING statement.

The LIMIT specification of the GDG definition limits the number of generations, instead of the FILENUM value in the TRACING statement. PowerExchange does not limit the number of GDG data sets. When the number of GDG data sets reaches the GDG LIMIT value, the operating system automatically removes the oldest generation from the GDG. If you specified the SCRATCH parameter when you created the GDG base, then these data sets are also deleted. For more information about defining GDG data sets, refer to the IBM DFSMS Access Method Services for Catalogs book.

Restriction: You cannot use DFSMS compression for alternative log data sets.


Review the following considerations before using DTLLOGnn DD statements to allocate alternative log data sets:

♦ You must specify the DTLLOG DD statement for the following purposes:

− To write initial startup messages

− To write messages if the DBMOVER configuration file member t does not specify a TRACING statement

− To write messages if PowerExchange encounters an error allocating or writing to alternative log data sets

The DTLLOG DD statement can point to a permanent data set or a SYSOUT data set. Informatica recommends that you specify use SYSOUT=* for this DD statement as a safeguard for any failures with alternative log files and to prevent the out-of-space conditions that can occur when you use permanent data sets.

♦ Unless you specify DTLLOGnn DD statements in the JCL, PowerExchange dynamically allocates log files when you use the TRACING statement.

♦ To prevent loss of PowerExchange log and trace information, use unique data sets for each PowerExchange job with DTLLOGnn DD statements.

♦ If you specify a GDG data set in any of the DTLLOGnn DD statements, PowerExchange uses the GDG data set and ignores other DTLLOGnn DD statements.

♦ PowerExchange jobs that write to the alternative log files includes the PowerExchange Listener, Netport jobs, PowerExchange Condense, and the PowerExchange Agent. The Adabas, Datacom, IDMS, and IMS log-based ECCRs can also use alternative log files.

The PowerExchange Logger, batch VSAM ECCR, CICS/VSAM ECCR, and IMS synchronous ECCR do not write messages to either the PowerExchange log or alternative log files.

Consumer API (CAPI) Connection Statements

PowerExchange requires CAPI connection statements to configure how change data is captured and extracted from the change stream for a database instance.

You must specify CAPI connection statements in the DBMOVER configuration file of the PowerExchange installation that will process the change stream. If you offload extraction processing, some additional configuration considerations apply. For more information, see the PowerExchange CDC Guide for Linux, UNIX, and Windows and PowerExchange CDC Guide for z/OS.

You must define at least one source-specific CAPI_CONNECTION statement for each source type. For MVS sources and for Oracle, you must also specify the UOWC CAPI_CONNECTION statement for the UOW Cleanser.

If necessary, you can specify multiple CAPI_CONNECTION statements of the same type or of different types in the same DBMOVER file.

Multiple CAPI Connections and Overrides

PowerExchange can capture changes for more than one source type using a single PowerExchange Listener and DBMOVER configuration file on a single server. You can define a maximum of eight CAPI_CONNECTION statements in a DBMOVER configuration file.

The CAPI_CONNECTION statements can be of the same type or of mixed types, as indicated by the TYPE parameter. The NAME parameter in each statement must specify a unique name.

If you define multiple CAPI_CONNECTION statements for a source type, you can optionally define a CAPI_SRC_DFLT statement to identify the default CAPI_CONNECTION for that source type. Also, you can optionally specify a CAPI_CONN_NAME parameter that specifies an overall default statement, out of all of the CAPI_CONNECTION statements defined in the DBMOVER file.


Instead of specifying defaults, you can use the following CAPI connection name overrides to point to a specific CAPI_CONNECTION statement:

♦ For a CDC session, the CAPI Connection Name Override attribute in the PowerCenter PWX application connection.

♦ For CAPXRT database row tests in the PowerExchange Navigator, the CAPI Connection Name value in the Advance dialog box. If you add a SQL statement for generating restart tokens, you can include the CONNAME parameter to point to the override CAPI_CONNECTION.

♦ If you use the DTLUAPPL to generate restart tokens, the CONN_OVR parameter in the DTLUAPPL control statement.

♦ For the PowerExchange Logger for Linux, UNIX, and Windows, the CONN_OVR parameter in pwxccl.cfg configuration file.

♦ For PowerExchange ODBC connections, the DTLCONN_OVR parameter in the odbc.ini file or the SQL escape sequence override DTLCONNOVR.

Informatica recommends that you use the following overrides:

♦ If you extract change data from Linux, UNIX, and Windows sources, specify a CAPI Connection Name Override value in the application connections for your CDC sessions.

♦ If you use the PowerExchange Logger for Linux, UNIX, and Windows, specify the CONN_OVR parameter in the pwxccl.cfg file.

The following example shows multiple source-specific CAPI_CONNECTION statements for DB2 and Oracle, with an overall CAPI_CONN_NAME default and a CAPI_SRC_DFLT default for DB2:

CAPI_CONN_NAME=DTLUDB1/* CAPI_CONNECTION=(NAME=DTLUDB1,TYPE=(UDB,CAPINAME=CAPIUDB1))CAPI_CONNECTION=(NAME=CAPIUDB1,TYPE=(UDB,DATABASE=SAMPLE1,DBCONN=SAMPLE1,PASSWORD=user,USERID=password,CCATALOG=DTLCCATLAOG))/* CAPI_CONNECTION=(NAME=DTLUDB2,TYPE=(UDB,CAPINAME=CAPIUDB2))CAPI_CONNECTION=(NAME=CAPIUDB2,TYPE=(UDB,DATABASE=SAMPLE2,DBCONN=SAMPLE2,PASSWORD=password,USERID=user,CCATALOG=DTLCCATLAOG))/*CAPI_CONNECTION=(NAME=DTLUDB,TYPE=(UDB,CAPINAME=CAPIUDB,))CAPI_CONNECTION=(NAME=CAPIUDB,TYPE=(UDB,DATABASE=SAMPLE,DBCONN=SAMPLE,PASSWORD=password,USERID=user,CCATALOG=DTLCCATALOG))CAPI_SRC_DFLT=(UDB,DTLUDB2)/*ORACLEID=(OEMCAP,OEMDB,OEMDB.INFORMATICA.COM,OEMDB.INFORMATICA.COM)CAPI_CONNECTION=(NAME=CAPIUOWC,TYPE=(UOWC,CAPINAME=CAPIORA))CAPI_CONNECTION=(NAME=CAPIORA,TYPE=(ORCL,ORACOLL=OEMCAP))

CAPI_SRC_DFLT Statements

You can specify a CAPI_SRC_DFLT statement for each source type to identify the default CAPI_CONNECTION statement for that source type.

The CAPI_SRC_DFLT statement has the following syntax:

CAPI_SRC_DFLT=(source_type,capi_connection_name)

The CAPI_SRC_DFLT must point to a CAPI_CONNECTION statement of a compatible type. The following table shows, for each CAPI_SRC_DFLT source type, the compatible CAPI_CONNECTION type:

CAPI_SRC_DFLT Type CAPI_CONNECTION Statement Type

AS4 UOWC

CAPX CAPX

Consumer API (CAPI) Connection Statements 43

For example, the following statements identify the default CAPI_CONNECTION statements for DB2 for Linux, UNIX, and Windows and for Oracle:

CAPI_SRC_DFLT=(UDB,DTLUDB2)CAPI_SRC_DFLT=(ORA,CAPIORA)

For more information about CAPI_SRC_DFLT, see “DBMOVER Parameter Descriptions” on page 4.

Order of Precedence for CAPI Statements and Overrides

PowerExchange uses the following order of precedence when determining which CAPI connection information to use:

1. CAPI connection name overrides take precedence over the CAPI_SRC_DFLT and CAPI_CONN_NAME statements.

2. A CAPI_SRC_DFLT statement for a specific source type takes precedence over a CAPI_CONN_NAME statement, with regard to that source type.

3. If neither a CAPI connection name override nor a CAPI_SRC_DFLT statement is specified, PowerExchange uses the CAPI_CONN_NAME statement.

4. If neither a CAPI connection name override nor CAPI_SRC_DFLT statement is specified for the source, and a CAPI_CONN_NAME statement is also not available, PowerExchange uses the first CAPI_CONNECTION statement for the source type in the DBMOVER configuration file.

Tip: Informatica recommends that you specify a CAPI connection name override for the most efficient CAPI processing.

Source-Specific CAPI_CONNECTION StatementsAll PowerExchange CDC platforms require CAPI_CONNECTION statements for capture and extraction processing. You define these statements in the DBMOVER configuration file. The types of CAPI_CONNECTION statements that you define vary by source type and platform. The following table identifies the required and optional CAPI_CONNECTION statements types by platform:

DB2, IMS, ADA, IDM, VSAM, VSM, DCM, IML or IDL UOWC

ORA UOWC

MSS MSQL

UDB UDB

Source Platform CAPI_CONNECTION Types Reference

All MVS Sources - LRAP CAPI_CONNECTION for the Log Read API that extracts change data from PowerExchange Logger for MVS log files (Required)

- UOWC CAPI_CONNECTION for the UOW Cleanser (Required)

- “LRAP CAPI_CONNECTION Parameters” on page 50

- “UOWC CAPI_CONNECTION Parameters” on page 60

DB2 for i5/OS - AS4J CAPI_CONNECTION for the journal reader that extracts change data fromDB2 journals (Required)


- “AS4J CAPI_CONNECTION Parameters” on page 45


DB2 for Linux, UNIX, and Windows

- UDB CAPI_CONNECTION for change extraction from DB2 recovery logs (Required)

- CAPX CAPI_CONNECTION, if you use continuous extraction mode (Optional)

- “UDB CAPI_CONNECTION Parameters” on page 58

- “CAPX CAPI_CONNECTION Parameters” on page 49

CAPI_SRC_DFLT Type CAPI_CONNECTION Statement Type


AS4J CAPI_CONNECTION Parameters

The AS4J CAPI_CONNECTION statement specifies the parameters needed for DB2 for i5/OS sources.

This statement has the following syntax:

CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(AS4J, [ALWCLRPFM=Y|N], [ALWPARTIAL=Y|N], [AS4JRNEXIT=Y|N], [EOF=Y|N], INST=instance_name, JOURNAL=journal, [LIBASUSER=Y|N], [NOCCUOWSZ=uow_size] [POLWAIT=secs], [STOPIT=(behaviour,exceptions)], [UOWRSTANY=Y|N] ))

Microsoft SQL Server - MSQL CAPI_CONNECTION for change extraction from SQL Server distribution databases (Required)


- “MSQL CAPI_CONNECTION Parameters” on page 51


Oracle - ORCL CAPI_CONNECTION for change extraction from Oracle redo logs (Required)



- “ORCL CAPI_CONNECTION Parameters” on page 53



Source Platform CAPI_CONNECTION Types Reference


The following table describes the parameters:

Parameter Required Default Description

ALWCLRPFM No N Controls whether DB2 for i5/OS CDC processing stops or continues when PowerExchange encounters changes that result from an i5/OS Clear Physical File Member (CLRPFM) command issued against a DB2 table registered for change data capture. PowerExchange cannot capture changes that result from a CLRPFM command.

Valid values are:- Y. PowerExchange ignores the CLRPFM command and

continues CDC processing. The data integrity of the CDC target might be damaged. If you also specify the AS400EVENTMSGQ statement in the DBMOVER configuration file, PowerExchange issues the message DTL3002 to the specified message queue when a journal entry for a CLRPFM command is encountered.

- N. PowerExchange CDC processing stops when changes from a CLRPFM command are detected.

Warning: If you set this parameter to Y, the data integrity of the CDC targets might be damaged. Specify this parameter only at the direction of Informatica Global Customer Support.

ALWPARTIAL No N Controls whether PowerExchange processes journal receivers in partial status or not.

Specify one of the following values:- Y. PowerExchange processes journal receivers in partial

status.- N. PowerExchange fails processing if a journal receiver is

in partial status.

Warning: If you specify Y for this parameter, you might compromise the data integrity of the change data being extracted because required changes might be unavailable. Only specify this parameter when instructed to do so by Informatica Global Customer Support.

AS4JRNEXIT No N Controls whether PowerExchange uses an exit program installed at the Delete Journal Receiver exit point, QIBM_QJO_DLT_JRNRCV, to prevent the deletion of journal receivers being processed or not.

Specify ones of the following values:- Y. PowerExchange locks any journal receiver that it is

processing so that it cannot be deleted from the i5/OS system. PowerExchange records journal receivers that it is processing in the CONDLIB library in a lock file called PWXJRNLCKP. When PowerExchange switches to the next journal receiver on the chain, PowerExchange removes the record for the previous journal receiver from the lock file.

- N. PowerExchange does not lock journal receivers that it processes.

Note: To use the AS400EVENTMSGQ parameter in the DBMOVER member, you must specify AS4JRNEXIT=Y.

Default is N.

For more information about the PowerExchange journal exit, see the PowerExchange CDC Guide for i5/OS.


DLLTRACE No none Specifies the name of the TRACE= statement that activates internal DLL tracing for this specific CAPI. Specify this parameter when instructed to do so by Informatica Global Customer Support.

EOF No N Controls whether PowerExchange stops change data extractions when the end-of-log is reached or not.

Specify one of the following values: - Y. PowerExchange stops change data extractions when

end-of-log is reached. - N. PowerExchange does not stop change data

extractions when end-of-log is reached.

Tip: Because this parameter affects all users of the CAPI_CONNECTION, use one of the following alternative methods for stopping change data extractions at end-of-log:- Specify 0 for the Idle Time attribute on the PWX

DB2i5OS CDC Real Time application connection. - Specify COLL_END_LOG=1 in the CAPTPARM

parameter for Condense jobs.- Specify WAITTIME=0 in the ODBC data source.

Default is N.

INST Yes none The INST parameter is a name of your choice.It must be the same name as that used in the collection identifier when creating your Registration Group in the Navigator. If you are running condense it must also match the value of DBID in CAPTPARM member. You can set INST to the database name, although this is not a requirement.

JOURNAL Yes none The journal from which PowerExchange extracts the change data, in the format library/journal_name.

LIBASUSER No N Indicates whether PowerExchange populates the DTL__CAPXUSER column with the user ID or the library and file name.

Specify one of the following values:- Y. For each change record, the DTL__CAPXUSER

column contains the library and file name to which the change was made.

- N. For each change record, the DTL__CAPXUSER column contains the name of the user that made the change.

Default is N.

NAME Yes No A unique name for the CAPI_CONNECTION statement.Maximum is 8 characters.



NOCCUOWSZ No none The size, in records, of the UOW that PowerExchange creates when reading change records from i5/OS journal receivers that were created without commitment control.

If you do not specify commitment control when making changes to DB2 for i5/OS tables, PowerExchange creates a UOW for each change record. Use this parameter to increase the size of the UOWs that PowerExchange creates. PowerExchange and PowerCenter process larger UOWs more efficiently. Larger UOWs also reduce commit activity on the targets of the change data.

PowerExchange ignores this parameter for change records created with commitment control. DB2 for i5/OS.


Tip: To specify unique UOW size values for CDC sessions, use the Minimum Rows Per commit attribute on the PWX DB2i5OS CDC Real Time application connection instead of this parameter.

POLWAIT No 10 Time, in seconds, that PowerExchange waits reaching the end of a journal receiver before checking for new change data in that journal receiver.Valid values are 0 through 10.

STOPIT=(behavior,exceptions)

No (CONT,5) Specifies how the extraction behaves when it encounters exceptions in the Journal.

An exception might be one of the following:- An After Image without a Before Image.- An attempt to delete a record that has no data.

The behavior option can be one of the following values:- CONT. Specifies that the extraction continues to run after

the specified number of exceptions.- TERM. Specifies that the extraction ends after the

specified number of exceptions.

The exceptions option specifies the number of exceptions, from 0 to n. The value 0 indicates that the extraction continues to run but does not report any exceptions.

TRACE No none Specifies the name of the TRACE= statement that activates the common CAPI tracing. Only specify this parameter when instructed to do so by Informatica Global Customer Support.



CAPX CAPI_CONNECTION Parameters

The CAPX CAPI_CONNECTION statement in dbmover.cfg specifies parameters for continuous extraction of change data from PowerExchange Logger log files. In continuous extraction mode, extractions run in near real time and read the data in the PowerExchange Logger log files as the change stream.

The CAPX CAPI_CONNECTION statement has the following syntax:

CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(CAPX, DFLTINST=collection_id, [FILEWAIT=nnnnn], [RSTRADV=sssss] ))


TYPE Yes none Type of CAPI_CONNECTION statement. For DB2 for i5/OS, this value must be AS4J.

UOWRSTANY No N Controls the point at which extracts can be restarted.

Specify one of the following values:- N. PowerExchange starts reading change data from the

journal receiver at a start-UOW boundary. If the restart point is not a start-UOW, PowerExchange fails the change data extraction.By failing processing, PowerExchange ensures data integrity by processing all change data for tables registered for capture.

- Y. PowerExchange can start reading change data at any point in the journal receiver. PowerExchange discards change data for all in-flight UOWs and issues a PWX-06734 warning message when the end-UOW is encountered. In-flight UOWs are those UOWs for which PowerExchange has not seen a start-UOW.

If you cannot guarantee a point at which there are no in-flight UOWs, you might need to specify UOWRSTANY=Y to allow change date extractions to be restarted.

Default is N. Warning: When you use this parameter, you might skip change data and compromise the data integrity of the targets of the change data.


DFLTINST Yes None A source identifier, sometimes called the instance name, that is defined in capture registrations. This value must match the instance or database name that is displayed in the Resource Inspector of the PowerExchange Navigator for the registration group that contains the capture registrations.Maximum length is eight characters.

DLLTRACE No None Name of the TRACE statement that activates internal DLL tracing for this specific CAPI.

Specify this parameter only if Informatica Global Customer Support instructs you to do so.



LRAP CAPI_CONNECTION Parameters

Use the LRAP CAPI_CONNECTION statement to specify parameters for the PowerExchange Logger Log-Read API. The Log-Read API retrieves captured change data by connecting to the PowerExchange Logger on MVS. The change data is then passed to the UOW Cleanser to be reconstructed into complete UOWs before being returned to the extraction.

Also specify a UOWC CAPI_CONNECTION statement. The CAPINAME parameter of the UOWC statement must specify the same value as the NAME parameter in the LRAP statement.

The LRAP CAPI_CONNECTION statement has the following syntax:

CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(LRAP, AGENT=agent_id, [EOF=Y|N], LOG=logger_id, [UIDFMT=ALL|CONN|CORR|CTYPE|PLAN|UID] )

FILEWAIT No 1 Number of seconds that PowerExchange waits before checking for new PowerExchange Logger log files.Valid values are from 1 through 86400.

NAME Yes None A unique name for the CAPI_CONNECTION statement.Maximum length is eight characters.

RSTRADV No None Number of seconds that PowerExchange waits before advancing the restart tokens for a data source by returning an empty unit of work (UOW). Empty UOWs contain restart tokens only, without any data. PowerExchange uses the restart tokens to determine the start point in the change stream for a PowerExchange extraction.

The wait period starts after a UOW for a data source is processed. PowerExchange resets the wait period when the next UOW is received or when an empty UOW is returned because the wait period expired, whichever comes first.

For example, if you specify 5 and a data source has no changes for 10 seconds, PowerExchange waits 5 seconds and then returns an empty UOW to advance the restart point for the source.

If RSTRADV is not specified, PowerExchange does not return empty UOWs to advance the restart point.

Valid values are from 0 through 86400.

Warning: A value of 0 can degrade performance because PowerExchange returns an empty UOW after every UOW processed.

TRACE No None Name of the TRACE statement that activates common CAPI tracing. Specify this parameter only if Informatica Global Customer Support instructs you to do so.

TYPE Yes None Statement type.For continuous extraction mode, this value must be CAPX.



)

MSQL CAPI_CONNECTION Parameters

Use the MSQL CAPI_CONNECTION statement to specify parameters for Microsoft SQL Server sources.


CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(MSQL, DISTDB=distribution_database, DISTSRV=distribution_server, [DWFLAGS=flag1flag2flag3], [EOF=Y|N], [MEMCACHE=cache_value],


AGENT Yes none Specify the PowerExchange Agent ID.The value specified must match the AGENTID value in the EDMSDIR module in the EDMPARMS DD statement of the PowerExchange Listener.Maximum is 4 characters.

DLLTRACE No none Specifies the name of the TRACE= statement that activates internal DLL tracing for this specific CAPI. Only specify this parameter when instructed to do so by Informatica Global Customer Support.

EOF No N Specifies whether PowerExchange ends extractions when end-of-log is reached (EOF=Y) or not (EOF=N). Since this parameter affects all users of the CAPI_CONNECTION, use one of the following methods instead:- Specify Idle Time=0 on the PWXPC Real Time application

connection. - Specify WAITTIME=0 in the ODBC data source.- Specify COLL_END_LOG=1 in the CAPTPARM parameter

for Condense jobs.

LOG Yes none Specify the 4 character PowerExchange Logger ID.The value specified must match the LOGGER value in the EDMSDIR module in the EDMPARMS DD statement of the PowerExchange Listener.

NAME Yes none Specifies a unique name for the CAPI_CONNECTION statement.Maximum is 8 characters.

TRACE No none Specifies the name of the TRACE= statement that activates the common CAPI tracing. Only specify this parameter when instructed to do so by Informatica Global Customer Support.

TYPE Yes none Specify LRAP.

UIDFMT No UID Specifies which fields PowerExchange returns in the DTL__CAPXUSER field:- UID - user identifier - PLAN - plan name - CORR - correlation identifier - CONN - connection identifier - CTYPE - connection type - ALL - all of the above If you specify ALL, the DTL__CAPXUSER field contains all of the information in a colon delimited list. The order of the values is as follows: UID:PLAN:CORR:CONN:CTYPE.Note: You can only specify one option. If you need more than one option, specify ALL.


[POLWAIT=ssss], [RSTRADV=sssss] ))


DISTDB Yes None Name of the SQL Server distribution database.

DISTSRV Yes None Network name of the SQL Server distribution server.

DLLTRACE No None Name of the TRACE statement that activates internal DLL tracing for this specific CAPI. Specify this parameter only when instructed to do so by Informatica Global Customer Support.

DWFLAGS No N, for all flags

A series of three positional parameters that determine whether processing stops or continues when data loss, truncation, or schema changes occur:- flag1 warns on data truncation. Enter Y to force processing to

continue if data of an unexpected length is received. Enter N to stop processing in this situation.

- flag2 warns on a schema change. Enter Y to force processing to continue if schema changes are detected. Enter N to stop processing in this situation.

- flag3 warns on the loss of change data. Enter Y to force processing to continue if the requested start sequence is not found in the transaction log. Enter N to stop processing in this situation.

Specify this parameter only when instructed to do so by Informatica Global Customer Support.

MEMCACHE No 248 KB Amount of memory cache, in kilobytes, that is allocated for caching a single row update.


NAME Yes None A unique name for the CAPI_CONNECTION statement.Maximum is eight characters.

POLWAIT No 10 Number of seconds the PowerExchange waits after reaching the end of current data before polling for new data. Valid values are from 1 through 2147483647.

RSTRADV No None Number of seconds PowerExchange waits before advancing the restart tokens for a data source by returning an empty unit of work (UOW). Empty UOWs contain restart tokens only, without any data. The restart tokens are used to determine the start point in the change stream for a PowerExchange extraction.

The wait period starts after a UOW for a data source is processed. PowerExchange resets the wait period when the next UOW is received or when an empty UOW is returned because the wait period expires, whichever comes first.

For example, if you specify RSTRADV=5 and a data source has no changes for 10 seconds, PowerExchange waits 5 seconds and then returns an empty UOW to advance the restart point for the source.



Warning: A value of 0 can degrade performance because PowerExchange returns an empty UOW after every UOW processed.


ORCL CAPI_CONNECTION Parameters

Use the ORCL CAPI_CONNECTION statement to specify parameters for Oracle sources.

Note: Specify a corresponding UOWC CAPI_CONNECTION statement. The CAPINAME parameter in the UOWC statement must specify the same value as the NAME parameter in the ORCL CAPI_CONNECTION statement.

The ORCL CAPI_CONNECTION statement has the following syntax:

CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(ORCL, [ARRAYSIZE=nnnn], [BYPASSUF=Y|N], [CATBEGIN=hh:mm], [CATEND=hh:mm], [CATINT=mins], [COMMITINT=mins], [GENRLOCK=Y|N], [LOGDEST=dest], [LGTHREAD=thread], [ONLINECAT=Y|N], ORACOLL=collection_id, [SELRETRY=retry], [SNGLINST=ora_instance] ))

TRACE No None Name of the TRACE statement that activates common CAPI tracing. Specify this parameter only when instructed to do so by Informatica Global Customer Support.

TYPE Yes None Type of CAPI_CONNECTION statement. For Microsoft SQL Serve, this value must be MSQL.


ARRAYSIZE No 100 Size of the prefetch array that PowerExchange uses to read the Oracle archive logs. Valid values are from 0 through 2147483647. Specify 0 only when instructed to do so by Informatica Global Customer Support to disable prefetch. Default is 100, which causes PowerExchange to request that Oracle prefetch 100 rows at a time.Note: A value of less than 100 can adversely affect Oracle capture performance.



BYPASSUF No N Indicates whether PowerExchange ends abnormally or issues a warning message whenever it receives an unformatted log record from Oracle LogMiner. LogMiner returns unformatted log records when Global Temporary Tables are updated, or when ONLINECAT=Y is specified and the log data being read is not consistent with the catalog. Valid values are:- N. PowerExchange ends with an error whenever it receives

an unformatted log record from Oracle LogMiner. - Y. PowerExchange issus a message that warns that

unformatted log data has been found and then continue CDC processing. Depending on the amount of unformatted log data, many warning messages can be issued. Specify IGNUFMSG=Y to suppress these messages.

Default is N.Tip: Specify Y if your Oracle instance contains Global Temporary tables. Otherwise, do not include the BYPASSUF parameter.

CATBEGIN No 00:00 Earliest 24-hour time of day at which Oracle can attempt to write the Oracle catalog to the Oracle archive logs. Enter this value in the format hh:mm.Either specify both CATBEGIN and CATEND with values other than their defaults, or specify neither parameter.

CATEND No 24:00 Latest 24-hour time of day at which Oracle can attempt to write the Oracle catalog to the Oracle archive logs. Enter this value in the format hh:mm.Either specify both CATEND and CATBEGIN with values other than their defaults, or specify neither parameter. If you specify neither parameter, the CATEND default of 24:00 is used.

CATINT No 1440 The interval, in minutes, between attempts to copy the Oracle catalog to the Oracle archive logs.If this interval elapses but the time is outside of the time period that is specified by the CATBEGIN and CATEND parameters, the Oracle catalog is not copied to the Oracle archive logs. The next attempt is made when the CATBEGIN time is reached.Valid values are from 1 through 1440.

COMMITINT No 5 Number of minutes between the Oracle COMMIT points that are generated by PowerExchange. PowerExchange uses Oracle LogMiner to read change data from Oracle redo logs. Even though PowerExchange does not update data in user tables, the Oracle LogMiner interface can generate transactions on behalf of PowerExchange. These transactions are written to the Oracle redo log. To restart efficiently, PowerExchange must occasionally issue SQL COMMIT calls. The COMMIT calls ensure that the transactions that Oracle LogMiner generated on behalf of PowerExchange terminate in a timely manner.

Without occasional COMMITs, the start point of the current real-time extraction or PowerExchange Logger session might become the restart point of the next extraction or PowerExchange Logger session, even if the sessions are unrelated. PowerExchange always begins reading data in the change stream from the start point of the oldest in-flight UOW. If this situation occurs, restart processing is prolonged because the data that has been logged since the start of the current session is re-read when the next session starts.Valid values are from 1 through 60.




GENRLOCK No N Indicates whether PowerExchange places exclusive locks on Oracle source tables while creating restart points for change data extraction. This parameter is used only in the following situations:- When the DTLUAPPL utility is used to create a non-default

restart token.- When the PowerExchange Logger for Linux, UNIX, and

Windows is cold started without the SEQUENCE_TOKEN and RESTART_TOKEN parameters because these parameters are not specified in the dtlca.cfg configuration file.

Specify one of the following values:- Y. DTLUAPPL or PowerExchange Logger places exclusive

locks on Oracle source tables so that PowerExchange can create valid restart points that do not skip change records for the tables. After the tables are locked, PowerExchange searches the Oracle catalog for the earliest active transaction to use as restart point, or for the current point in the log if an active transaction does not exist. While PowerExchange is determining restart points, transactions that contain updates for the tables that PowerExchange has locked cannot begin. After the restart points are generated, PowerExchange releases the locks on the source tables, which allows the tables to be updated again.

- N. PowerExchange creates restart points for source tables based on the current point in the redo log. These restart points might cause PowerExchange to skip transactions that contain updates for the tables if these transactions were active at the time the restart points were created.Specifying N is the same as not including the GENRLOCK parameter.

Default is N.

IGNUFMSG No N If you specify BYPASSUF=Y, you can specify this parameter to indicate whether PowerExchange issues a warning message whenever unformatted data is found and then continues CDC processing.

Specify one of the following values:- N. PowerExchange does not issue the warning messages.- Y. PowerExchange issues the warning messages.Default is N.Note: This parameter is ignored if you specifyBYPASSUF=N.



LOGDEST No None Numeric identifier for the archive log destination that you want to force PowerExchange to use. This archive log destination must be local to the Oracle instance where PowerExchange extraction sessions and the PowerExchange Logger run.

For example, to use archived logs from the destination set by the LOG_ARCHIVE_DEST_3 parameter in the init.ora file, specify LOGDEST=3.


The SNGINST parameter affects how PowerExchange uses the archive log destination specified by LOGDEST.- If SNGINST=N, or if SNGINST is not specified,

PowerExchange uses the specified archive destination to search for archived redo logs that contain copies of the Oracle catalog. After PowerExchange passes these logs to an Oracle LogMiner session, LogMiner determines the other archived redo logs to read.

- If you specify SNGLINST=Y, you must define both LOGDEST and LGTHREAD to specify the only archive log destination and Oracle instance from which LogMiner reads archived redo logs. LogMiner does not read any other archived redo logs. After the logs at the specified location have been processed, the extraction session ends.

If you specify ONLINECAT=Y, LOGDEST and LGTHREAD are validated but ignored.

LGTHREAD No None Instance number of the Oracle instance that you want to force PowerExchange to use for identifying the archived redo logs to process. - In a RAC environment, you can use this parameter to force

PowerExchange to use the archived logs that belong to a specific Oracle instance in the RAC.

- In a non-RAC environment, enter 1. A value other than 1 causes PowerExchange to not find the redo log from which to restart change data extraction.


The SNGINST parameter affects how PowerExchange uses the Oracle instance specified by LGTHREAD.- If SNGINST=N, or if SNGINST is not specified,

PowerExchange uses the specified Oracle instance to search for archived redo logs that contain copies of the Oracle catalog. After PowerExchange passes these logs to an Oracle LogMiner session, LogMiner determines the other archived redo logs to read.

- If you specify SNGLINST=Y, you must define both LOGDEST and LGTHREAD to specify the only archive log destination and Oracle instance from which LogMiner reads archived redo logs. LogMiner does not read any other archived redo logs. After the logs at the specified location have been processed, the extraction session ends.

If you specify ONLINECAT=Y, LOGDEST and LGTHREAD are validated but ignored.

NAME Yes None A unique name for the CAPI_CONNECTION statement.Note: This value must match the CAPINAME parameter value in the a UOWC CAPI_CONNECTION statement.

Maximum length is eight characters.

ORACOLL Yes None The Oracle collection identifier that is specified in the ORACLEID statement.



ONLINECAT No N Indicates whether PowerExchange directs Oracle LogMiner to use the Oracle online catalog or the copy of the catalog in the redo logs to format log data for CDC.

Specify one of the following values:- N. Oracle LogMiner uses the copy of the catalog from the

archived redo logs. With this setting, PowerExchange can track schema changes to ensure that data loss does not occur.

- Y. Oracle LogMiner uses the online catalog. With this setting, PowerExchange CDC initializes faster. However, because LogMiner does not track DDL changes, LogMiner cannot format log records for tables that have schema changes. If you make schema changes while LogMiner is reading log data, LogMiner passes unformatted log records to PowerExchange for the changes that are recorded after the schema changed. The consequences depend on the BYPASSUF parameter. If you use BYPASSUF=N (the default), the extraction request fails when Oracle passes the first unformatted record. If you use BYPASSUF=Y, PowerExchange skips the unformatted record and continues processing, which results in data loss.

Note: If you specify ONLINECAT=Y, you still must copy the online catalog to the Oracle redo logs on a routine basis. PowerExchange requires this catalog copy to determine the point from which to restart change data extractions.

Restrictions: Do not set ONLINECAT=Y under the following conditions:- You need to start an extraction from a point in the Oracle

redo logs that contains table data that was captured under a previous schema.

- You set BYPASSUF=Y and need to change the schema for tables that are registered for CDC while change data extraction is occurring.

SELRETRY No 1000 Number of times that PowerExchange immediately loops back to the Oracle LogMiner call before implementing a graduated-scale wait loop.

After this number of successive retries of a call to LogMiner, PowerExchange implements a wait interval between each subsequent retry. The wait interval begins at one millisecond and gradually increases to one second. When LogMiner returns data, the wait interval is reset to 0, and the process begins again for the next call to LogMiner.


If you specify a non-zero value, PowerExchange uses non-blocking SQL to ensure that a user request to shut down a extraction session is processed in a timely manner.

If you specify 0, PowerExchange does not use non-blocking SQL. This setting improves CPU consumption but can prolong extraction session shutdown. On quiescent Oracle instances, PowerExchange does not honor a shutdown request until log data is returned from Oracle. On Oracle instances where update activity is occurring, shutdown behavior does not noticeably change.



UDB CAPI_CONNECTION Parameters

Use the UDB CAPI_CONNECTION statement to specify parameters for DB2 for Linux, UNIX, and Windows sources.


CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(UDB, [CCATALOG=capture_catalog], [DBCONN=dbname], [EPWD=encryted_password], [MEMCACHE=cache_value], [PASSWORD=password], [RSTRADV=sssss], [SPACEPRI=primary], [UDBSCHEMA=schema], [UPDINT=sssss], [USERID=user_id] ))

SNGLINST No N Indicates whether to force PowerExchange to use only the archived redo logs from a specific Oracle instance and archive log destination.

If you specify Y, you must also specify the LOGDEST and LGTHREAD parameters to identify the archive log destination and Oracle instance to use. For the other Oracle instances in the RAC, you must run separate change data extraction processes. Then determine how to properly merge changes for loading to a target.

Note: In an Oracle 9i RAC environment, PowerExchange cannot perform CDC unless you set this parameter to Y.


TYPE Yes Type of CAPI_CONNECTION statement. For Oracle, this value must be ORCL.


CCATALOG No creator.DTLCCATALOG

Where creator is the user ID that is used to connect to the database.

Name of the PowerExchange capture catalog table in the format creator.table_name.

DBCONN No None Database name that overrides the database name in the extraction map. You can use this parameter to extract change data to another identical database using the original capture registrations.




EPWD No None An encrypted password. You can create encrypted passwords by using the PowerExchange Navigator.

If you specify the USERID parameter, you must specify either the PASSWORD or EPWD parameter. Do not specify both PASSWORD and EPWD.

MEMCACHE No 1024 KB Amount of memory cache, in kilobytes, that is allocated to reconstruct complete UOWs.

PowerExchange keeps all changes in each UOW in cache until it completes reading the end-UOW (commit record). If the MEMCACHE value is too small to hold all of the changes in a UOW in cache, the changes spill to a disk file.


NAME Yes None A unique name for the CAPI_CONNECTION statement.

Maximum length is eight characters.

PASSWORD No None A clear text password.

If you specify the USERID parameter, you must specify either the PASSWORD or EPWD parameter. Do not specify both PASSWORD and EPWD.

RSTRADV No None Number of seconds that PowerExchange waits before advancing the restart tokens for a data source by returning an empty unit of work (UOW). Empty UOWs contain restart tokens only, without any data. PowerExchange uses the restart tokens to determine the start point in the change stream for an extraction.

The wait period starts after a UOW for a data source is processed. The wait period is reset when the next UOW is received or when an empty UOW is returned because the wait period expires, whichever comes first.

For example, if you specify RSTRADV=5 and a data source has no changes for 10 seconds, PowerExchange waits five seconds and then returns an empty UOW to advance the restart point for the source.



Warning: A value of 0 can degrade performance because PowerExchange returns an empty UOW after every UOW that is processed.


TYPE Yes None Type of CAPI_CONNECTION statement. For DB2 for Linux, UNIX, and Windows, this value must be UDB.

UDBSCHEMA No None A schema name that overrides the schema name in capture registrations.



UOWC CAPI_CONNECTION Parameters

Use the UOWC CAPI_CONNECTION statement to specify parameters for the UOW Cleanser. In the change stream, changes from multiple UOWs are intermingled. The UOW Cleanser reconstructs the UOWs from the change stream into complete and consecutive UOWs that are in chronological order based on end time.

The UOWC CAPI_CONNECTION statement has the following syntax:

CAPI_CONNECTION=( NAME=name, [DLLTRACE=trace_id], [TRACE=trace], TYPE=(UOWC, CAPINAME=capiname, [BLKSIZE=block_size], [DATACLAS=data_class], [MEMCACHE=cache_value], [RSTRADV=sssss], [SPACEPRI=primary], [SPACETYPE=BLK|TRK|CYL], [STORCLAS=storage_class], [UNIT=unit] ))

UPDINT No 600 Minimum number of seconds that PowerExchange must wait after encountering a virtual timestamp (VTS) in the DB2 log records for a partition before writing a postioning entry to the PowerExchange capture catalog table. The positioning entry, which is composed of a log sequence number (LSN) and VTS, indicates a location in the DB2 logs. Note: The UPDREC minimum number of records must also be met before positioning entries can be written to the capture catalog table.

UPDREC No 10000 Minimum number of DB2 log records that PowerExchange must read for a partition before it can write a positioning entry to the PowerExchange capture catalog table. The positioning entry, which is composed of a LSN and VTS, indicates a location in the DB2 logs.Note: The UPDINT minimum wait period must also be met before positioning entries can be written to the capture catalog table.

USERID No None User ID for access to the DB2 database.If you specify this parameter, you must also specify either the PASSWORD or EPWD parameter.Note: This user ID must have SYSADM or DBADM authority.


BLKSIZE No i5/OS: 32760MVS: 18452Oracle: 32768

Block size, in bytes, for the sequential UOW spill file. Valid values are:- 8 through 32760 for MVS and i5/OS platforms- 8 through 65535 for Oracle

CAPINAME Yes None The value that is specifed for the NAME parameter in the source-specific CAPI_CONNECTION statement, which can be an AS4J, LRAP, or ORCL CAPI_CONNECTION statement.

DATACLASS No None SMS data class to use for the temporary sequential UOW spill data sets. This parameter applies to MVS only.




MEMCACHE No 1024 KB (1 MB)

Amount of memory cache, in kilobytes, that is allocated to the UOW Cleanser for reconstructing complete UOWs.The UOW Cleanser keeps all changes for each UOW in cache until the end-UOW (commit record) is read. If the memory cache is too small to hold all of the changes in a UOW, the UOW Cleanser spills the changes to a sequential disk file, called the UOW spill file.


NAME Yes None A unique name for the CAPI_CONNECTION statement.Maximum length is eight characters.

RSTRADV No None Number of seconds PowerExchange waits before advancing the restart tokens for a data source by returning an empty UOW. Empty UOWs contain restart tokens only, without any data. The restart tokens are used to determine the start point in the change stream for a PowerExchange extraction.

The wait period starts after a UOW for a data source is processed. The wait period is reset when the next UOW is received or when an empty UOW is returned because the wait period expires, whichever comes first.

For example, if you specify 5 and a data source has no changes for 10 seconds, PowerExchange waits 5 seconds and then returns an empty UOW to advance the restart point for the source.


Warning: A value of 0 adversely affects performance. This setting causes PowerExchange to return an empty UOW after every UOW processed.


SPACEPRI No 50 cylinders Size to use for allocating a UOW spill file on MVS. When the file becomes full, another spill file of the same size is allocated. The type of size unit is specified by the SPACETYP parameter. ACS routines can override this size.

Note: On i5/OS, spill files are allocated as physical files with SIZE(*NOMAX). This option causes the maximum spill file size to be determined by the system maximum file size. On Linux, UNIX, and Windows, spill files are temporarily allocated in the capture directory with a size of 2 GB.

SPACETYP No BLK Type of units in which space is allocated for the sequential UOW spill data sets. Valid values are:- BLK for blocks- CYL for cylinders- TRK for tracksThis parameter applies to MVS only.

STORCLASS No None SMS storage class name to use for the temporary sequential UOW spill data sets.This parameter applies to MVS only.

TRACE No None Name of the TRACE statement that activates common CAPI tracing. Specify this parameter only when instructed to do so by Informatica Global Customer Support..



ICU-specific Configuration Parameters

The following DBMOVER configuration parameters are specific to International Components for Unicode (ICU):

TYPE Yes None Type of CAPI_CONNECTION statement. For the UOWC Cleanser, this value must be UOWC.

UNIT No SYSDA Generic or esoteric unit name for the temporary sequential UOW spill data sets.This parameter applies to MVS only.

Parameter Description

ICUALIAS Syntax:ICUALIAS=(CPN,command,alias)Where:- CPN. A free code page number slot in the range 301 to 340. See the

ICUCHECK report to identify free slots.- command. ADD, DELETE or DELETEALL. The DELETEALL command only

uses two sub-parameters.- alias. The alternative name which can be used to find the code page. DB2

CCSID s are found using aliases beginning with "cp".

Examples:ICUALIAS=(302,ADD,CP930)Add an alias of CP930 which points at CPN 302. This means that DB2 columns with CCSID 930 will use this custom ICU code page.

ICUALIAS=(206,DELETE,CP930)The alias of "CP930" is deleted from CPN206.

ICUALIAS=(206,DELETEALL)All aliases are deleted from CPN206.

ICUCNVPROPERTY Syntax:ICUCNVPROPERTY=(CPN,property,value)Where: - CPN. A free code page number slot in the range 301 to 340. See the

ICUCHECK report to identify free slots- property. A value from the following list. If you specify an invalid property, the

full list of properties is displayed in the error messages.- value. Can be ON or OFF for the binary properties, or a numeric value for

buffer_increment.



ICUCNVPROPERTY continued

The following values are valid property values:- STATE_BYTES. Causes two extra bytes to be allocated when deciding the

buffer size when converting to the code page.- UNKNOWN_ENDIANNESS. Flags that the code page has different integer

"endianness" depending on the platform. In some situations the system can determine the correct code page (for example switch from UTF-16 to UTF-16LE on Windows) but in general it is best to never use such code pages.

- ASCII. Indicates that the characters 0-9 and A-Z match the 7 bit ASCII values, such as in code page US_ASCII.

- EBCDIC. Indicates that the characters 0-9 and A-Z match the standard EBCDIC values. For example, in code page IBM-37.

- SQL_PARSABLE. Indicates that the invariant SQL characters match the standard values for the ASCII or EBCDIC type. This is used to determine whether the ODBC layer needs to translate SQL before looking for literals like SELECT, FROM, or WHERE.

- LOWERCASE_NONSTANDARD. There are 4 Japanese MVS code pages where the lower case a-z characters do not match the standard EBCDIC value used by IBM-37. SQL for these code pages is forced into upper case in the client program before it is converted to the MVS code page. CP930 uses this property.

- FIX_MIN_LENGTH. ICU processing has a limitation that it drops leading binary zeros when converting to double-byte code pages. Setting this property ensures that all characters are created at the correct minimum length with leading binary zeros inserted when necessary. The Show a code page uses this property.

- BUFFER_INCREMENT. Causes extra bytes to be allocated when deciding the buffer size when converting to the code page. The value should be a number.

- POWERCENTER_NAME. Defines a PowerCenter name by which the code page can be found. The name should match an entry in the list box from Workflow Manager -> define relational sources. On Windows, ICU code pages must have a PowerCenter name to be used with ODBC.

Notes:- The ICUCNVPROPERTY parameter should only need to be specified in rare

situations. - The properties default to OFF or zero on new custom ICU code pages.Examples:ICUCNVPROPERTY=(301, STATE_BYTES, ON)ICUCNVPROPERTY=(301, UNKNOWN_ENDIANNESS, ON)ICUCNVPROPERTY=(301, ASCII, ON)ICUCNVPROPERTY=(301, EBCDIC, ON)ICUCNVPROPERTY=(301, SQL_PARSABLE, ON)ICUCNVPROPERTY=(301, LOWERCASE_NONSTANDARD, ON)ICUCNVPROPERTY=(301, FIX_MIN_LENGTH, ON)ICUCNVPROPERTY=(301, BUFFER_INCREMENT, 8)ICUCNVPROPERTY=(301, POWERCENTER_NAME, " IBM EBCDIC Japanese")



pwxcmd Command Configuration Statements

To configure a PowerExchange Listener or a PowerExchange Logger for Linux, UNIX, and Windows process to receive pwxcmd commands, you must perform the following configuration tasks:

1. Configure and start the PowerExchange processes to which you want to send pwxcmd commands.

2. Configure connections to the PowerExchange processes from any PowerExchange environments from which you will issue pwxcmd commands.

The following table describes the statements that you use to configure PowerExchange processes to receive pwxcmd commands:

ICUCONVERTER Syntax:ICUCONVERTER=(CPN,command,code_page_name,min_bytes, max_bytes,space_hex)Where:- CPN. A free code page number slot in the range 301 to 340. See the

ICUCHECK report to identify free slots- command. ADD or DELETE. The DELETE command uses only two

subparameters- code_page_name. The name of the code page converter. This is also a

component of the CNV file name- min_bytes. Minimum bytes per character- max_bytes. Maximum bytes per character- space_hex. Hexadecimal value of the space character. The system checks that

this character can be converted in UTF-8 during initialization. The character is appended to CHAR columns where the data is short.

Examples:ICUCONVERTER=(301,ADD,user_showa_01,2,2,0000)Adds a code page called user_showa_01 into CPN 301. All characters have 2 bytes and the space character is \X00\X00.ICUCONVERTER=(206,DELETE)Deletes a code page and all its aliases from the system. The only reason for doing this would be if a replacement custom ICU code page was being added.ICUCONVERTER=(302,ADD,user_cp930,1,2,40)Adds a code page called user_cp930 into CPN 302. Characters use either 1 or 2 bytes and the space character is \X40.

ICUDATADIR Defines the directory where the binary CNV files produced by makeconv are located. Note that on all platforms except MVS, the directory name is case sensitive.Windows example:ICUDATADIR=c:\ucm\ShowaUcmIf the parameter is present, then Power Exchange validates all custom ICU code pages by opening them and converting the space character into UTF-8. This validation fails if:- The binary CNV file cannot be loaded (U_FILE_ACCESS_ERROR)- The space character did not convert into (any character) in UTF-8ICUDATADIR is mandatory if you specify ICUCONVERTER ADD parameters.

Statement Configuration File Defines Reference

On the machine where the PowerExchange process runs:

CONDENSENAME pwxccl.cfg file A name for the command-handling service for a PowerExchange Logger for Linux, UNIX, and Windows process that receives pwxcmd commands

PowerExchange CDC Guide for Linux, UNIX, and Windows



SVCNODE StatementThe SVCNODE statement specifies the TCP/IP port on which a command-handling service for a PowerExchange Listener or PowerExchange Logger for Linux, UNIX, and Windows process listens for pwxcmd commands.

Syntax

The syntax for this statement is:

SVCNODE=(service_name,port_to_listen_on)

Parameters

The parameters for this statement are:

♦ service_name is either:

− For a PowerExchange Listener process, the node name specified in the LISTENER statement in the DBMOVER configuration file. This LISTENER statement configures the PowerExchange Listener to which you want to send pwxcmd commands.

− For a PowerExchange Logger for Linux, UNIX, and Windows process, the service name specified in the CONDENSENAME statement in the pwxccl.cfg file. For information about the CONDENSENAME statement, see the PowerExchange CDC Guide for Linux, UNIX, and Windows.

♦ port_to_listen_on is a unique port number on which the command-handling service for the PowerExchange process listens for pwxcmd commands.

For examples of SVCNODE statements, see the PowerExchange Command Reference.

CMDNODE StatementThe CMDNODE statement specifies connection information for a PowerExchange Listener or PowerExchange Logger for Linux, UNIX, and Windows process to which you want to send pwxcmd commands.

Syntax

The syntax for this statement is:

CMDNODE=(service_name,service_type,host_name,port_to_connect_to)

LISTENER dbmover.cfg file A PowerExchange Listener on the current platform

“DBMOVER Parameter Descriptions” on page 4

SVCNODE dbmover.cfg file The TCP/IP port on which a command-handling service for a PowerExchange process listens for pwxcmd commands

“SVCNODE Statement” on page 65

On the machine from which you issue pwxcmd commands:

CMDNODE dbmover.cfg file Connection information for a PowerExchange process to which you want to send pwxcmd commands

“CMDNODE Statement” on page 65

Statement Configuration File Defines Reference

pwxcmd Command Configurat ion Statements 65

Parameters

The parameters for this statement are:

♦ service_name is the name of the command-handling service for the PowerExchange process to which you want to send pwxcmd commands. This name can be any name you want. Use this service name when issuing pwxcmd commands to this PowerExchange process.

♦ service_type is either:

− LISTENER. A PowerExchange Listener process.

− CONDENSE. A PowerExchange Logger for Linux, UNIX, and Windows process.

♦ host_name is the host name or IP address of the PowerExchange process to be contacted.

♦ port_to_connect_to is the port number on which the command-handling service for the PowerExchange process listens for pwxcmd commands. This port number must match the port number in the corresponding SVCNODE statement.

For examples of CMDNODE statements, see the PowerExchange Command Reference.

Example DBMOVER Configuration Files

PowerExchange provides an example DBMOVER configuration file for each platform. Use the example files as a starting point when creating your configuration files.

The example files are available at the following locations:

♦ On i5/OS, the DBMOVER member in the the CFG file in the datalib library

♦ On Linux, UNIX, and Windows, the dbmover.cfg_sample file in the PowerExchange installation directory

♦ On z/OS, the DBMOVER member in the RUNLIB library

Note: The z/OS DBMOVER member includes variables in pointed brackets < >, which are replaced with specific values when you complete the MVS Installation Assistant.


C H A P T E R 3

Netport Jobs


♦ Netport Jobs Overview, 67

♦ Configuring for Netport, 67

♦ Netport Jobs Description, 68

♦ PowerExchange Substitution Variables for Netport Jobs, 68

♦ Netport Jobs and Offload Processing, 70

♦ Netport Jobs and DB2 Bulk Data Loads, 70

♦ Using Netport Jobs for Generation Data Sets, 70

Netport Jobs Overview

PowerExchange uses parameters in the DBMOVER configuration file to link JCL streams called Netport jobs to a port. Netport jobs are useful in some situations. If you use Netport jobs, consider the following points:

♦ Access must be through a specific non-PowerExchange module, such as IMS access where DLIBATCH or IMS BATCH are executed.

♦ Excessive wait times might be experienced. For example, if data is on tape, a significant delay might result from waiting for tape mounts.

♦ If you want to use GDGs with Netport jobs, special considerations and configuration requirements apply. For more information, see “Using Netport Jobs for Generation Data Sets” on page 70.

♦ If you want to access IDMS data when full user ID checking is required, such as when the SECURITY parameter in the DBMOVER configuration file is set to 2,x.

Configuring for NetportTo configure a Netport job:

1. Assign a port to listen for Netport job requests.

2. Associate a JCL stream to that port.

67

The DBMOVER configuration file configures a Netport job for tapes and GDGs and another Netport job for IMS, unless IMS is not required. The following parameters are included at installation time for the IMS Netport:

LISTENER=(node1,TCPIP,12480)NETPORT=(node1,12480,,,"DTLUSR.V800B11.RUNLIB(IMSJCL)",PSB1)

To run a row test from the PowerExchange Navigator to read IMS data, you need to add a node to the dbmover.cfg file on Windows similar to the following:

NODE=(imsnet,TCPIP,12.34.56.78,12480)

When running an IMS row test against a data map, use the node name (“imsnet” in the previous example) to access the data. The PowerExchange Listener will cause the IMSJCL job to be submitted, passing the PSB (PSB1) name to the JCL as a parameter.

For any new Netport job, a new PowerExchange Listener NODE statement and an associated NETPORT statement are required. Use the LISTENER and NETPORT statements in the TAPEJCL member as a template. In the NETPORT statement, assign a suitable port to the Netport job. Make sure that the positional commas remain.

The first parameter in both the LISTENER and NETPORT statements (“node1” in the previous example) must point to the node name in the LISTENER statement for the port on which this PowerExchange Listener is listening. The default configuration uses node1 and 2480.

Netport Jobs Description

Netport jobs are provided in the following members: the TAPEJCL member of the RUNLIB library for tape access and the IMSJCL member for IMS access. If you need other Netport jobs, create them using the JCL in the TAPEJCL member as a template. The JCL in the TAPEJCL member runs a PowerExchange module, whereas the JCL in the IMSJCL member runs an IMS module.

If IDMS access is required and the SECURITY parameter in the DBMOVER configuration file is set to 2 (each user who submits an IDMS access job must have their user ID and password checked by the relevant security package), you must use a Netport job. Copy the TAPEJCL member under a new and relevant name. Make sure that the PowerExchange copies of the IDMS.LOADLIB and IDMS.DBA.LOADLIB libraries are in the STEPLIB. Code the following DD statement in the JCL that you created:

//SYSIDMS DD DSN=&HLQ..RUNLIB(DTLDMCL),// DISP=(SHR)

Ensure that the relevant SYSCTL statement is included in the JCL for Central Versions, or the following DD statements if running local:

//IDMSDCT INCLUDE MEMBER=IDMSDICT//IDMSFIL INCLUDE MEMBER=IDMSFILE

These members will need the relevant dictionary definitions and database file definitions. Edit the DBMOVER configuration file as specified in “Configuring for Netport” on page 67.

Note: The PowerExchange Listener must be restarted to detect a new Netport job.

PowerExchange Substitution Variables for Netport Jobs

The fifth positional parameter of the NETPORT statement in the DBMOVER configuration file specifies the skeleton JCL to use the Netport job. To control and customize Netport jobs, PowerExchange provides substitution variables. Prior to submitting the Netport job to MVS, PowerExchange resolves any PowerExchange substitution variables found in the skeleton JCL with the appropriate values.

68 Chapter 3: Netport Jobs

PowerExchange provides sample JCL in the NETJCL member of the RUNLIB library, which demonstrates how you can use substitution variables to form unique data set names.

The following table describes the PowerExchange substitution variables for Netport jobs:

Substitution Variable Name

Description

%CLIENT Unique 16-byte hexadecimal identifier that includes the job name of the PowerExchange Listener

%CPNODE PowerExchange Listener node name to which the connection to the PowerExchange Listener was made

%CPPORT PowerExchange Listener port on which the connection to the PowerExchange Listener was made

%DATE_YYMMDD Current Gregorian date

%DATE_YYYYDDD Current Julian date

%Nn Value of the numeric counter with the length of n. Valid values for the length are 1 through 7.

The PowerExchange Listener maintains a separate value for each possible numeric counter, and increases the value by one each time that numeric counter is used. For example, if Netport JCL specifies %N3 in two places in the JCL, the first occurrence is assigned a value of 001 and the second occurrence is assigned a value of 002. The next Netport job that references %N3 is assigned 003, and so on.

%PWD The password of the user ID that connected to the PowerExchange Listener. To avoid exposing passwords, you should only use this substitution variable on the PASSWORD parameter of the JOB card.

%SOCKNO Socket number on which the connection to the PowerExchange Listener was made

%SMFTASK If the STATS statement in the DBMOVER configuration file specifies recording to SMF, the PowerExchange task number. Otherwise, the value is 00000.

%TIME_HHMMSS Current time, in 24 hour clock time format

%USER The user ID that connected to the PowerExchange Listener

%1 Value of the sixth positional parameter from the NETPORT statement in the DBMOVER configuration file

%2 Value of the seventh positional parameter from the NETPORT statement in the DBMOVER configuration file

%3 Value of the eighth positional parameter from the NETPORT statement in the DBMOVER configuration file

%4 Value of the ninth positional parameter from the NETPORT statement in the DBMOVER configuration file

%5 Value of the tenth positional parameter from the NETPORT statement in the DBMOVER configuration file

%6 Value of the eleventh positional parameter from the NETPORT statement in the DBMOVER configuration file

%7 Value of the twelfth positional parameter from the NETPORT statement in the DBMOVER configuration file

%8 Value of the thirteenth positional parameter from the NETPORT statement in the DBMOVER configuration file

PowerExchange Substitution Variables for Netport Jobs 69

Netport Jobs and Offload Processing

If you are using Netport jobs and offload processing together, make sure that the PowerExchange Listener and the Netport jobs point to the same data map file. This data map is specified using the //DATAMAP DD statements in the JCL of the PowerExchange Listener and the Netport jobs.

Netport Jobs and DB2 Bulk Data Loads

When a DB2 bulk data load operation is run from a Netport job, the load runs as a task of the Netport job, not as a separately submitted job.

A sample Netport job, such as CAPXJCL, can be used as a template for creating such a Netport job. The sample job must be edited to add the DD statements that would normally be in the DB2 LOAD JCL.

For example:

//SORTOUT DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//SORTWK01 DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//SYSDISC DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//SYSERR DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//SYSMAP DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//SYSUT1 DD UNIT=SYSDA,SPACE=(4000,(20,20),,,ROUND)//UTPRINT DD SYSOUT=*

Note: The Netport job cannot have a DD statement for SYSPRINT. This output is allocated to a file for DB2 LOAD.

Using Netport Jobs for Generation Data Sets

MVS Generation Data Groups (GDGs) offer a unique and versatile approach to accessing data. Using a relative generation number, you can reference the most current generation data set in the GDG using a fixed data set name. This practice is extremely useful for PowerExchange data maps and partner ETL tools. The GDG name can be put in a single data map without requiring dynamic updates to the data map data set name. You can also write to GDG data sets using the relative generation number.

The ProblemWhen you run a batch job and access a GDG data set using a relative generation number, such as AAA.BBB.CCC(0), you access the same generation data set for the life of the batch job regardless of whether more current generations exist. This is a good way to handle things for MVS batch jobs. With long running batch jobs or with started tasks such as the PowerExchange Listener, you normally want to access the current generation data set when using relative references.

The SolutionBy default, the PowerExchange Listener does not refresh the generation table after the first reference to a generation data set. After the PowerExchange Listener accesses a generation data set using a relative generation number, all future references to that GDG access the same generations as the first reference. As a result, you cannot reference any newly created generations.


To have the PowerExchange Listener refresh the generation table when using relative generation references, choose one of the following options:

♦ Enter GDGLOCATE=Y in the DBMOVER configuration member. This setting will ensure that the catalog is referenced when access to a GDG is requested (reading an existing data set or creating a new data set) to make the latest information available. This feature prevents the need for recycling the PowerExchange Listener.

♦ Set up a NETPORT job for GDG types to allow the latest GDGs to be referenced. This practice is similar to how IMS is handled. Each time the file is accessed, the NETPORT job shuts down and a subsequent call is made to invoke another NETPORT job. All GDGs are recognized as you intended.

NETPORT jobs are used for IMS, IDMS, and for accessing tapes with long mount times.

Ensure that the NETPORT job runs on the same MVS image as the PowerExchange Listener to satisfy socket API call constraints. Otherwise, the jobs might timeout.

♦ Recycle the PowerExchange Listener at regular intervals, more frequently than the GDGs are created or need to be created by means of PowerExchange.

Setting Up Netport Jobs for GDGs

To set up a NETPORT Job for GDGs:

1. Determine an available MVS port number (“7777” in this example) to use with the Netport job.

2. Add a NODE statement to your source and target DBMOVER configuration files, for example:

NODE=(MVS,TCPIP,mvs1,2480)NODE=(MVSGDG,TCPIP,mvs1,7777) ******

If you are using ODBC drivers to point from partner software, you need to add a driver for this new location.

3. Update the MVS DBMOVER configuration file in the RUNLIB library:

LISTENER=(node1,TCPIP,2480)LISTENER=(node1,TCPIP,7777)

To indicate to the MVS PowerExchange Listener that the second port is not typical, add a NETPORT statement with the appropriate node name and port number (in this case “node1” and port “7777”), for example:

NETPORT=(node1,7777,,,"INFA.Vxyz.RUNLIB(GDGJCL)",,,,,,,,)

4. Create a member in the RUNLIB library called GDGJCL, using the TAPEJCL member as a template.

When the MVS PowerExchange Listener detects activity on the Netport, instead of directly trying to service it, it tries to submit the JCL file named in the fifth token of the NETPORT statement to the JES internal reader. As it is reading and submitting the JCL, it scans every JCL line for tokens %x (where x is a numeric value of 1 to 8). It substitutes these tokens with the appropriate value taken from parameter 6 to 13 of the NETPORT statement. Do not confuse the %x tokens with the %Nx tokens. The %Nx tokens are used to generate incremented numeric values of x length. (In the example JCL below, this value is used on the job card to generate unique job names.)

To create the GDGJCL job from the TAPEJCL job, add a job card (leaving "%N5" appended to it) and any other DDs that you might need for your PowerExchange Listener job. The GDGJCL job can then be submitted whenever a request is received through the assigned port for the Netport job. This enables a job to end and another job to begin and handle the GDGs on MVS appropriately.

5. Stop and restart the MVS PowerExchange Listener.

Whenever you access a GDG, point to the MVSGDG location instead of the normal MVS.

Using Netport Jobs for Generation Data Sets 71


C H A P T E R 4

Message Destination Override


♦ Overview, 73

♦ DTLMSGO File, 74

Overview

The text for the majority of PowerExchange messages is contained within the DTLMSG file, which is created during the installation process using the contents shipped with the product. PowerExchange displays these messages in several possible locations:

♦ PowerExchange log file (DTLLOG or an alternative log file)

♦ Standard output (stdout)

♦ System operator console (MVS and i5/OS only)

Using the message destination override capability, you can determine to which, if any, of these locations a particular message is routed. This capability is useful for redirecting messages to destinations other than their default ones when you want to enable automation or adhere to installation-specific output handling standards. This capability also enables you to suppress specific messages. Before suppressing messages, consider the implications. Usually, destination overrides are used for only a small number of messages.

To configure Message Destination Override, create a file called DTLMSGO. In this file, specify the message number followed by the destinations to which the message is to be written. The manner in which the location of the DTLMSGO file is specified to PowerExchange varies by platform. For more information, see “DTLMSGO File” on page 74.

Note: On MVS, messages other than those in the DTLMSG file can be issued. These messages are composed of the messages written by the following PowerExchange components using the EDMMSG DD statement:

♦ PowerExchange Agent (a small number of messages are issued from the DTLMSG file)

♦ PowerExchange Logger

♦ Log-Read API and Log-Write API

♦ Environmental Change Capture Routines (ECCRs), including the batch VSAM, CICS/VSAM, Datacom, DB2 for z/OS, IDMS synchronous, and IMS synchronous ECCRs

73

DTLMSGO File

The location of the DTLMSGO file varies by platform:

♦ On MVS, the JCL must contain a DTLMSGO DD statement that points to a sequential data set or PDS member. A sample file is provided in the DTLMSGO member of the RUNLIB library.

♦ On i5/OS, the DTLMSGO member resides in the dtllib/DTLMSG file.

♦ On Linux, UNIX, and Windows, the dtlmsgo.txt file resides in the PowerExchange installation directory. A sample file, called dtlmsgo_sample.txt, is provided in the installation directory.

Syntax Rules for DTLMSGO StatementsUse the following syntax rules when defining DTLMSGO statements:

♦ Comment lines must begin with “/*” in column 1.

♦ If the same message appears more than once in the DTLMSGO file, the last occurrence is used.

Syntax for DTLMSGO StatementsUse the following syntax for DTLMSGO statements:

msg_number,{msg_dest},{msg_dest}, ... | msg_number

Required Parametermsg_number

This is the message number from the DTLMSG file (dtlmsg.txt on UNIX/Windows). The message numbers are 5-digit numbers with no prefix.

No generic or masking capability exists so complete message numbers must be specified, although it is not necessary to specify the leading zeroes.

Optional Parametermsg_dest

Valid values for this parameter are:

♦ null. If no destination is specified, the message is not rerouted to a different destination. Rather, the message is treated as if it is not included in the DTLMSGO file.

♦ LOG. Specifies that the message is directed to the PowerExchange Log. This is either the DTLLOG file or the alternative logging file. If alternative logging is used, the message might be redirected to the DTLLOG file if it is issued prior to the alternative logging subtask being initialized.

♦ NONE. Specifies that the message is suppressed. The message is not displayed in any of the other locations. This parameter is mutually exclusive with other msg_dest values.

♦ STDOUT. Specifies that the message is directed to the standard output location, which varies by platform.

On MVS, STDOUT is sent to the SYSPRINT DD if specified or a dynamically allocated temporary JES SPOOL data set if not specified.

On i5/OS, STDOUT is sent to QPRINT if PowerExchange is running as a background job or to the terminal if is running interactively.

On UNIX and Windows, STDOUT is sent to the terminal by default unless you have redirected it elsewhere.

74 Chapter 4: Message Destination Override

♦ SYSCONS. On MVS, specifies that the message is displayed by the Write-To-Operator (WTO) macro. No routing code is associated with the WTO so the ROUTCODE value specified in the DEFAULT statement of CONSOLxx is used for routing.

On i5/OS, SYSCONS results in the message being sent to the operator message queue.

On Linux, UNIX, and Windows, SYSCONS is ignored.

DTLMSGO File ExampleThe following example DTLMSGO file shows how to code the following types of messages: messages without leading zeroes (594 and 607), messages without any change in destination (00595), and messages with multiple destinations (00650, 00651, and 607):

/*/* Sample file to change message destinations/*/* region size msg to console594,SYSCONS/* use pwx defaults00595/* multiple destinations00650,SYSCONS,LOG00651,SYSCONS,LOG,STDOUT/* suppress stats msg00408,NONE/*607,STDOUT,LOG/*

DTLMSGO File 75

76 Chapter 4: Message Destination Override

C H A P T E R 5

Statistics Logging Including SMF


♦ Overview, 77

♦ Configuring for SMF on z/OS, 78

♦ Configuring for File Output, 78

♦ Record Format, 78

Overview

PowerExchange delivers the following types of records:

♦ Connection record driven by the PowerExchange Listener

♦ Start record driven by the PowerExchange process, Netport, PowerExchange Listener, and subtask

♦ Interval record for all processes

♦ End record for all processes (normal and abnormal end)

The following information is collected:

♦ General connection information

♦ General resource measurements

♦ Access method specific information

♦ Access method specific measurements

The record number to which this information is written is specified in the DBMOVER configuration file. When writing data to SMF, choose this record number in consultation with your system programmers or capacity planning team. Statistics logging is available for use on z/OS writing data to SMF and for testing purposes writing data to a flat file.

Note: The interval statistics show the cumulative statistics for all tasks that are active at a given time. These statistics are useful when monitoring the activity and progress of long running tasks. By examining the statistics in two interval reports for a specific task, you can determine all of the activities of the task within a given time interval.

Interval logging reports only on active tasks. Consequently, the total relates only to tasks that are active when the interval statistics are reported. Interval logging does not report on the following types of activities:

♦ Any activity by tasks that started and completed between the two interval reports

♦ Any activity by a task that terminated since the last interval report

77

Configuring for SMF on z/OS

To perform SMF logging, specify the following STATS statement in the DBMOVER configuration file and then restart the PowerExchange Listener. By default, statistics are not collected (STATS=NONE).

STATS=(SMF,record_number,interval)

For example:

STATS=(SMF,255,0)


Note: To enable PowerExchange to write to SMF, all libraries in the STEPLIB of the PowerExchange Listener must be APF-authorized.

Configuring for File Output

For testing purposes, it is possible to write statistics to file output rather than to SMF. This practice is not suitable for long-term statistics collection. However, you might want to occasionally use this method to get data that you need on the same day as you request it. To begin statistics logging to file, specify the following STATS statement in the DBMOVER configuration file and then restart the PowerExchange Listener. By default, statistics are not collected (STATS=NONE).

STATS=(FILE,filename,interval)

For example:

STATS=(FILE,DTLUSR.V800.SMFFILE,0)

To allow the formatted data to be read easily, a data map, called pwxstat.file.dmp, is shipped with PowerExchange. Use this data map to read statistics gathered using the FILE subparameter. You can find pwxstat.file.dmp in the following location: PowerExchange_install_directory\examples\datamaps.

Warning: On MVS, use this type of statistics logging only for short-term testing or debugging. Writing directly to a file, whether a single sequential file or GDG, can cause operational problems for the PowerExchange environment. For z/OS logging, use the format described in “Configuring for SMF on z/OS” on page 78.

Record Format

When logging statistics to SMF, the record has a standard SMF header, including Sub Type identifier. File output also includes an SMF header for consistency. However, the detail included in this record is limited. All

Parameter Default Description

record number none SMF record number. Range is 128 to 255.

interval 0 Interval in minutes. Valid values are 0 and any number in the range 5 to 120. An interval of 0 denotes no interval processing.

Parameter Default Description

filename none File name of the file to which the statistic should be written.

interval 0 Interval in minutes. Valid values are 0 and any number in the range 5 to 120. An interval of 0 denotes no interval processing.

78 Chapter 5: Statistics Logging Including SMF

formats include a PowerExchange header containing triplets that define occurrences, lengths and offsets for all included record sections:

Standard SMF Header with Sub Type IncludedThe following table shows the standard SMF header with the subtype included:

PowerExchange Header/DescriptorThe following table shows the PowerExchange header/descriptor:

Sub Type Content

1 General section only

2 General, function/access method

3 General, extended, function/access method




OffsetHex Offset

Field Length Format Description

0 0 SMFxLEN 2 Binary Record length

2 2 SMFxSEG 2 Binary Segment descriptor

4 4 SMFxFLG 1 Binary Header flag byte

5 5 SMFxRTY 1 Binary Record type

6 6 SMFxTME 4 Binary Time, in hundredths of a second, that it took to move the record to the SMF buffer

10 A SMFxDTE 4 Packed Date on which the record was moved to SMF buffer, in the format 0CYYDDDF, where F is the sign byte

14 E SMFxSID 4 EBCDIC System ID (taken from the SID parameter)

18 12 SMFxSSI 4 EBCDIC SUBSYS=option

22 16 SMFxSTY 2 Binary Record Subtype

OffsetHex Offset

Field Length Format Value/Description

0 0 Eye catcher 8 Char PWXHDR _

8 8 Clock time (start) 8 Binary

16 10 Clock time (end) 8 Binary

24 18 No of Extended Section 4 Binary

28 1C Offset to 1st Extended Section

4 Binary

Record Format 79

Extended Section DescriptorThe following table shows the extended section descriptor:

OffsetHex Offset


0 0 Section type 4 Binary 1. PowerExchange Listener Section.2. Exception Section.3. File Section.4. DB2 Section.5. Client Section.

4 4 Section Count 4 Binary Count of following sections of same type

8 8 Length of section 4 Binary


General SectionThe following table shows the general section:

Extended Section for PowerExchange Listener

OffsetHex Offset


0 0 Sub type 2 Binary x'0001'. Connect Request.x'0002'. Start.x'0003'. Interval.x'0004'. End.x'0013'. PowerExchange Listener Interval.lx'0014'. PowerExchange Listener End Same as in standard SMF header.

2 2 Product 14 Char 'PowerExchange '

16 10 Version/Build 10 Char Result=dtlinfoGetBuildVersion()

26 1A Component 16 Char 'Listener', utility name and so on

42 2A Job Name 8 Char PowerExchange Listener or process

50 32 Node name 8 Char When applicable

58 3A Process ID 5 Char 1-99999

63 3F Reserved 3 Binary Future use/alignment

66 42 Client Session ID

8 Char Client session ID. This value is passed through from the SESSID parameter of the connection parameters or the DtlSessid ODBC Escape Sequence.

74 4A Client IP addr 4 Binary From pucw

78 4E Client user id 128 Char From dbcb

206 CE Start Time 20 TIMESTAMP

YYYYMMDDHHMMSSNNNNNN(DB2 format)

226 E2 Reserved 2 Binary Alignment

228 E4 End Time2 20 TIMESTAMP

YYYYMMDDHHMMSSNNNNNN(DB2 format)

248 F8 Reserved 2 Binary Alignment

250 FA Return Code 8 Char

258 102 Reason Code 8 Char

266 10A Additional Info 24 Char For abnormal end/ non-zero RC

290 122 Reserved 32 Binary For future use

318 13E CPU Time 8 Binary CPU time, in microseconds

326 146 Descriptors length

4 Binary Length of descriptor sections

OffsetHex Offset


0 0 Job/STC name 8 Char Job or started task name

8 8 Nodename 8 Char Node name

16 10 Ip Address 16 Char

Record Format 81

Extended Section for PowerExchange Client

32 20 Start time 20 Char YYYYMMDDHHMMSSNNNNNN(DB2 format

52 34 End time 20 Char YYYYMMDDHHMMSSNNNNNN(DB2 format

72 48 STCK start 8 Binary Start time (STCK)

80 50 STCK end 8 Binary End time (STCK)

88 58 Port Number 4 Binary

92 5C Number connections

4 Binary Number of connections

96 60 Reserved 4 Binary

100 64 Number accepted connections

4 Binary Number of accepted connections

104 68 Number connections

4 Binary Number of connections refused due to memory exceeded

108 6C Number connections

4 Binary Number of connections refused due to task limit

112 70 CPU time 8 Binary CPU time of task, in microseconds

120 78 Bytes sent 4 Binary Bytes sent

124 7C Messages sent 4 Binary

128 80 Bytes received 4 Binary

132 84 Messages received

4 Binary

OffsetHex Offset


0 0 STCK start 8 Binary Start time (STCK)

8 8 STCK end 8 Binary End time (STCK)

16 10 CPU time 8 Binary CPU time of task, in microseconds

24 18 Bytes sent 4 Binary Bytes sent

28 1C Messages sent 4 Binary

32 20 Bytes received 4 Binary

36 24 Messages received 4 Binary

40 28 RC 1 4 Binary Return code

44 2C RC 2 4 Binary Return code

48 30 RC 3 4 Binary Return code

OffsetHex Offset



Extended Section for Data Access

Function/Access Method Section (DB2)

OffsetHex Offset


0 0 AM Type 1 Char

1 1 Reserved 7 Binary For future use

18 12 Rows Read 4 Binary

22 16 Bytes Read 8 Binary

30 1E Rows Written 4 Binary

34 22 Bytes Written 8 Binary

42 2A Access method 20 Char Name of access method

OffsetHex Offset


0 0 DB2 Time 8 Binary CPU time, in microseconds

8 8 No. Statements

4 Binary Number of DB2 statements executed

12 C No. Rows 4 Binary Number of rows returned

16 10 SSID 4 Char DB2 SSID

20 14 DB2 Plan Name

8 Char

28 1C Connection Type

8 Char RRSAF/CAF

36 24 Authorization id

8 Char

44 2C Correlation id 32 Char Correlation ID has the format PWX99999, where 99999 is a unique number generated by the PowerExchange Listener. A new Correlation ID will be generated for each subtask generated by the PowerExchange Listener.

76 4C SQL Code 4 Binary

80 50 Reason Code 4 Binary

84 54 Updated 4 Binary Rows updated

88 58 Deleted 4 Binary Rows deleted

92 5C Inserted 4 Binary Rows inserted

Record Format 83


C H A P T E R 6

PowerExchange Security


♦ PowerExchange Security Overview, 85

♦ SECURITY Statement in the DBMOVER Configuration File, 86

♦ i5/OS Security, 87

♦ Linux, UNIX, and Windows Security, 89

♦ MVS Security, 91

♦ PowerExchange Selective Sign-on, 100

♦ PowerExchange Sign-on File, 101

PowerExchange Security Overview

PowerExchange provides security options that you can use to authenticate users that attempt to connect to a PowerExchange Listener and to authorize access to resources by PowerExchange jobs and tasks. PowerExchange provides security options that pertain to all operating systems. Some security options are specific to an operating system or data source type.

PowerExchange provides the following security options:

♦ On all operating systems, PowerExchange provides selective sign-on capability. You can use selective sign-on to authorize connection access to PowerExchange. When a PowerExchange Listener accepts a new connection over TCP/IP, PowerExchange checks the PowerExchange sign-on file to verify access for the user ID and, optionally, the IP address.

♦ On i5/OS, PowerExchange can use operating system facilities to authenticate user IDs and passwords for users that connect to PowerExchange. PowerExchange can also use the supplied user IDs to check authority to access resources that PowerExchange jobs and tasks need to use.

♦ On Linux, UNIX, and Windows, PowerExchange can use operating system facilities to authenticate user IDs and passwords for use of the pwxcmd program. PowerExchange can also use the sign-on file to authorize user access to specific pwxcmd commands. On Windows, PowerExchange can use the sign-on file to authorize use of specific PowerExchange Listener commands issued through the PowerExchange Navigator.

♦ On MVS, PowerExchange can use operating system facilities to authenticate user IDs and passwords for users that connect to PowerExchange. PowerExchange can also use the supplied user IDs to check authority to access resources that PowerExchange jobs and tasks need to use. Additionally, PowerExchange provides source-specific security options for Adabas, Datacom, DB2, and IMS.

The SECURITY statement in the DBMOVER configuration file determines the level of security that PowerExchange provides.

85

SECURITY Statement in the DBMOVER Configuration File

Use the SECURITY statement in the DBMOVER configuration file to define the following types of security:

♦ User authentication to access PowerExchange

♦ User access to resources such as files and data sets by PowerExchange jobs and tasks on MVS and i5/OS

♦ User access to PowerExchange commands executed by the pwxcmd program on Linux, UNIX, and Windows

♦ User access to PowerExchange Listener commands executed by the PowerExchange Navigator on Windows.

The SECURITY statement has the following syntax:

SECURITY=({0|1|2},{N|Y})

The type and scope of the user authentication that PowerExchange performs depends on the value specified for the first parameter and on the operating system. On MVS and i5/OS, the first parameter controls the use of operating system user authentication and resource access. On Linux, UNIX, and Windows, the first parameter controls the use of operating system user authentication and access to pwxcmd commands. For more information, see the following topics:

♦ “i5/OS Security” on page 87

♦ “Linux, UNIX, and Windows Security” on page 89

♦ “MVS Security” on page 91

The second parameter controls the use of PowerExchange selective sign-on, and access on Windows to PowerExchange Listener commands issued through the PowerExchange Navigator. For more information, see “PowerExchange Selective Sign-on” on page 100.

The following table describes SECURITY settings on i5/OS and MVS:

SECURITY Statement

Description

SECURITY=(0,N)SECURITY=(0,Y)

- PowerExchange does not require users to specify a valid operating system user ID and password.

- All resource access uses the user ID of the PowerExchange Listener or PowerExchange Condense job.

The value of the second parameter determines the use of PowerExchange selective sign-on:- N. PowerExchange does not use selective sign-on.- Y. PowerExchange uses selective sign-on to control connections to PowerExchange.


- PowerExchange requires users to specify a valid operating system user ID and password.

- All resource access uses the user ID of the PowerExchange Listener or PowerExchange Condense job.



- PowerExchange requires users to specify a valid operating system user ID and password.

- All resource access uses the user ID supplied on the connection.


86 Chapter 6: PowerExchange Security

The following table describes SECURITY settings on Linux, UNIX, and Windows:

i5/OS Security

Use the SECURITY statement in the DBMOVER configuration file on i5/OS to configure the following types of PowerExchange security:

♦ User authentication. If you specify 1 or 2 in the first parameter of the SECURITY statement, PowerExchange uses operating system facilities to authenticate users that connect to and use PowerExchange. If you also configure PowerExchange selective sign-on, PowerExchange checks operating system user IDs and passwords after successful selective sign-on checking.

SECURITY Statement

Description


- PowerExchange does not require users to specify a valid operating system user ID and password.

- All resource access, excluding RDBMS resources, uses the user ID of the PowerExchange Listener or the PowerExchange Logger for Linux, UNIX, and Windows.

- RDBMS security controls RDBMS access.

The value of the second parameter determines the use of PowerExchange selective sign-on:- N. PowerExchange does not use selective sign-on.- Y. PowerExchange uses selective sign-on to control connections to PowerExchange,

and access on Windows to PowerExchange Listener commands issued through the PowerExchange Navigator.


- When you use the pwxcmd program, PowerExchange requires a valid operating system user ID and password. Otherwise, a user ID is not required.






- When you use the pwxcmd program, PowerExchange requires a valid operating system user ID and password. Otherwise, a user ID is not required.

- PowerExchange controls access to specific pwxcmd commands by checking the AUTHGROUP and USER parameters in the signon.txt file.





Note: When you specify 2 for the first parameter, PowerExchange uses the sign-on file for checking access to pwxcmd commands. The value of the second parameter does not affect this processing.

i5/OS Security 87

♦ Resource access. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses connection user IDs to authorize access to any resources that PowerExchange uses. Otherwise, resource access is controlled by the user ID under which PowerExchange jobs and tasks execute.

♦ Selective sign-on. If you specify Y in the second parameter of the SECURITY statement, PowerExchange uses the selective sign-on file to limit the users that connect to PowerExchange. Otherwise, any operating system user ID can connect to PowerExchange. For more information, see “PowerExchange Selective Sign-on” on page 100.

Resource Access Requirements for i5/OSWhen you specify 2 for the first parameter on the SECURITY statement, PowerExchange uses the connection user ID to control access to resources. The user ID must have access to specific PowerExchange resources and to any resources that PowerExchange uses.

To use PowerExchange, users require the following authority levels:

♦ To write messages to the PowerExchange log file, users must have *CHANGE authority to the datalib library. By default, PowerExchange create the datalib library with a default public authority of *USE.

♦ To read data maps, users must have *USE authority to the data map library. By default, this library is called STDATAMAPS. To specify a different name, set the DMX_DIR parameter in the DBMOVER member of the CFG file.

♦ To add, delete, or change data maps, users must have *CHANGE authority.

♦ To read and write to other libraries, files, and DB2 tables through PowerExchange, users must have the appropriate authority.

Access Requirements for PowerExchange Jobs

To perform bulk data movement and CDC operations, PowerExchange jobs and tasks require certain authority

The following table lists the access authority that the PowerExchange Listener requires for bulk data processing:

The following table lists the access authority that the PowerExchange Listener and PowerExchange Condense require for bulk data and CDC processing:

Library and File Name Authority Required

datalib/AUTHTSKLST1 *USE

datalib/AUTHTSKSTP1 *USE

datalib/CFG *USE

datalib/LOG *CHANGE

dtllib/DTLMSG *USE

dtllib/LICENSE *USE

stdatamaps *CHANGE

1 PowerExchange uses these files to control access to PowerExchange Listener commands. During the installation process, PowerExchange creates these files with a default public authority of *EXCLUDE. To allow a user to list or stop PowerExchange tasks, grant *USE authority to the user ID.

File NamePowerExchange Listener Authority

PowerExchange Condense Authority

cndlib/CFGCOND *USE *USE

cndlib/PWXJRNLCKP *CHANGE *CHANGE

cpxlib *CHANGE n/a


PowerExchange dynamically creates other objects, such as capture registrations and data maps. For users other than the user ID under which the PowerExchange Listener and PowerExchange Condense run, the default authority to access these objects depends on the following authority settings:

♦ The Create Authority system value, QCRTAUT, which determines the system-wide public authority for new objects

♦ The default public authority for the library that contains the objects

Access Requirements for Journals and Files

The user ID under which the PowerExchange Listener and PowerExchange Condense run must have the appropriate authority level to access journals and files that PowerExchange needs to use.

The following table lists the authority requirements by object type:

Linux, UNIX, and Windows Security

Use the SECURITY statement in the DBMOVER configuration file on Linux, UNIX, and Windows to configure the following types of PowerExchange security:

♦ User authentication for the pwxcmd program. If you specify 1 or 2 in the first parameter of the SECURITY statement, PowerExchange uses operating system facilities to authenticate users of the pwxcmd program. If you also configure PowerExchange selective sign-on, PowerExchange checks operating system user IDs and passwords after successful selective sign-on checking.

datalib/AUTHTSKLST1 *USE *USE

datalib/AUTHTSKSTP1 *USE *USE

datalib/CCT *CHANGE *USE

datalib/CDCT *USE *CHANGE

datalib/CDEP *CHANGE *CHANGE

datalib/CFG *USE *USE

datalib/LOG *CHANGE *CHANGE

dtllib/DTLMSG *USE *USE

dtllib/LICENSE *USE *USE

stdatamaps *CHANGE n/a

1 PowerExchange uses these files to control access to PowerExchange Listener commands. During the installation process, PowerExchange creates these files with a default public authority of *EXCLUDE. To allow a user to list or stop PowerExchange tasks, grant *USE authority to the user ID.

Object Authority

Journal *OBJEXIST

Journal library *EXECUTE

Journal receivers *USE

Library that contains journal receivers *EXECUTE

Files *USE

Library that contains files *EXECUTE

File NamePowerExchange Listener Authority


Linux, UNIX, and Windows Security 89

♦ Authorization for running pwxcmd commands. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses the PowerExchange sign-on file to check user authority to run specific pwxcmd commands. Otherwise, authority to run pwxcmd commands is not checked.

♦ Selective sign-on. If you specify Y in the second parameter of the SECURITY statement, PowerExchange uses the selective sign-on file to limit the users that connect to PowerExchange, and to check user authority to issue PowerExchange Listener commands through the PowerExchange Navigator on Windows. Otherwise, any user ID can connect to PowerExchange, and issue PowerExchange Listener commands through the PowerExchange Navigator on Windows. For more information, see “PowerExchange Selective Sign-on” on page 100.

User Authentication for the pwxcmd ProgramYou can configure PowerExchange to authenticate operating system user IDs and passwords to control the use of the pwxcmd program. To verify user credentials for pwxcmd program use, you must perform the following configuration tasks for each PowerExchange installation that is the target of a command from the pwxcmd program:

♦ Specify 1 or 2 for the first parameter of the SECURITY statement in the dbmover.cfg file.

♦ Configure the necessary permissions to authenticate operating system user credentials.

Authenticating User Credentials on Linux and UNIX

For PowerExchange to authenticate user credentials on Linux and UNIX systems, the operating system must use shadowed passwords. PowerExchange reads user IDs and passwords from the /etc/passwd and /etc/shadow file. By using the crypt function, PowerExchange then verifies that the passwords supplied on the pwxcmd command matches the password for the user ID.

Restriction: PowerExchange authentication of user credentials does not support Linux and UNIX systems with security that is not backwardly compatible with the shadow password file and crypt function, such as HP1 for HP-UX.

1. Change the owner of pwxauth.exe to root.

2. Change the group of pwxauth.exe to the group of the PowerExchange Listener.

3. Set the setuid bit on pwxauth.exe.

PowerExchange provides a sample script called setup_pwxauth in the PowerExchange installation directory that makes the necessary changes to pwxauth.exe. To run this script, issue the following command:

setup_pwxauth listener_primary_group

For the variable listener_primary_group, specify the group name of the PowerExchange Listener. The PowerExchange Listener should be the only UID in the group specified.

Authenticating User Credentials on Windows

To verify user credentials on Windows machines, PowerExchange uses the LogonUser API.

On Windows 2000 and earlier, the PowerExchange Listener user ID must have the SE_TCB_NAME privilege. In User Manager, this privilege is the “Act as part of the Operating System” right. On Windows Server 2003 and later and on Windows XP and later, no additional Windows configuration is necessary.

For more information about the LogonUser API, see the following Microsoft article “How to validate user credentials on Microsoft operating systems” at http://support.microsoft.com/kb/180548.

Restriction: When you enable security for the pwxcmd program on Windows, PowerExchange does not accept guest as an authorized user account.


Authorization for Running pwxcmd CommandsIn addition to authenticating user credentials for pwxcmd program use, you can configure PowerExchange to check authorization to run pwxcmd commands. If you specify 2 for the first parameter of the SECURITY statement on the machine that is the target of the command, PowerExchange checks the PowerExchange sign-on file to determine whether to allow the user ID supplied on the pwxcmd program to run commands.

In the PowerExchange sign-on file, you can create authorization groups, which PowerExchange uses in conjunction with user definitions to authorize users to run one or more pwxcmd commands. You must configure the sign-on file on each PowerExchange installation that is the target of command from the pwxcmd program. For more information, see “PowerExchange Sign-on File” on page 101.

Note: For PowerExchange to authorize users to run pwxcmd commands, you must have configured PowerExchange to check user credentials for the pwxcmd program. For more information, see “User Authentication for the pwxcmd Program” on page 90.

MVS Security

Use the SECURITY statement in the DBMOVER configuration file on MVS to configure the following types of PowerExchange security:

♦ User authentication. If you specify 1 or 2 in the first parameter of the SECURITY statement, PowerExchange uses operating system facilities to authenticate users that connect to and use PowerExchange. If you also configure PowerExchange selective sign-on, PowerExchange checks operating system user IDs and passwords after successful selective sign-on checking.

♦ Resource access. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses connection user IDs to authorize access to any resources that PowerExchange uses. Otherwise, resource access is controlled by the user ID under which PowerExchange jobs and tasks execute.

♦ Capture registration and extraction map access. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses FACILITY class profiles to control user access to capture registrations and extraction maps. Otherwise, the external security product controls access to capture registrations and extraction maps at the data set level only.

♦ Data maps access. If you specify 2 in the first parameter of SECURITY statement and specify Y for the DM_SUBTASK parameter in the DBMOVER configuration file, PowerExchange uses FACILITY class profiles to control user access to data maps. Otherwise, the external security controls access to data maps at the data set level only.

♦ PowerExchange Listener commands. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses FACILITY class profiles to control user access to PowerExchange Listener commands issued from the PowerExchange Navigator or the DTLUTSK utility. Otherwise, PowerExchange does not control commands issued from the PowerExchange Navigator or the DTLUTSK utility.

♦ Adabas file access. In PowerExchange data maps, you can specify passwords for Adabas files. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses FACILITY class profiles to control write access to Adabas files. Otherwise, PowerExchange does not control write access to Adabas files.

♦ Datacom table access. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses FACILITY class profiles to control read access to Datacom tables. Otherwise, PowerExchange does not control read access to Datacom tables.

♦ DB2 for z/OS access. If you specify 2 in the first parameter of the SECURITY statement and RRSAF in the MVSDB2AF parameter in the DBMOVER configuration file, PowerExchange uses the user ID from the PowerExchange connection to access DB2 resources. Otherwise, PowerExchange uses the user ID under which the PowerExchange Listener runs.

♦ IMS database access. If you specify 2 in the first parameter of SECURITY statement, PowerExchange uses FACILITY class profiles to control write access to IMS databases.Otherwise, PowerExchange does not control write access to IMS databases.

MVS Security 91

♦ PowerExchange Agent access. If you specify YES for the InitAuthCheck parameter in the AGENTCTL parameter file, PowerExchange uses FACILITY class profiles to control access to PowerExchange Agent services and commands. For more information about security for the PowerExchange Agent, see the PowerExchange CDC Guide for z/OS.

♦ Selective sign-on. If you specify Y in the second parameter of the SECURITY statement, PowerExchange uses the selective sign-on file to limit the users that connect to PowerExchange. Otherwise, any operating system user ID can connect to PowerExchange. For more information, see “PowerExchange Selective Sign-on” on page 100.

Resource Access Requirements on MVSWhen you specify 2 for the first parameter on the SECURITY statement, PowerExchange uses the user ID for the connection to PowerExchange for control access to resources. The user ID must have access to specific PowerExchange resources and to any resources that PowerExchange uses.

To use PowerExchange, users must have the following general levels of access:

♦ Users must have READ access to all PowerExchange data sets.

♦ If the PowerExchange Listener uses a data set for the PowerExchange message log file specified in the DTLLOG DD statement, users must have UPDATE access to that file.

♦ To create, update, or delete data maps, users must have UPDATE access to the VSAM data set specified in the DATAMAP DD statement of the PowerExchange Listener JCL.

♦ Appropriate authority to any data sets, databases, or tables that bulk data movement operations use.

Data Set Access Requirements for PowerExchange Jobs

The user ID under which PowerExchange jobs and started tasks run must have the proper authority to access the data sets that PowerExchange allocates for bulk data movement and CDC. PowerExchange jobs and started tasks use DD statements in their JCL to allocate data sets. They also dynamically allocate data sets during execution.

The following table lists the access authority that the PowerExchange Listener and Netport jobs require for bulk data processing:

The following table lists the access authority that PowerExchange jobs and started tasks require for bulk data movement and CDC processing:

DD NamePowerExchange Listener Authority

Netport Jobs Authority

STEPLIB READ READ

DATAMAP UPDATE READ

DTLCFG READ READ

DTLKEY READ READ

DTLLOG UPDATE UPDATE

DTLLOGnn UPDATE UPDATE

DTLMSG READ READ

DTLSGN READ READ


PowerExchange Agent Authority

PowerExchange Logger Authority


STEPLIB READ READ READ READ

DATAMAP UPDATE N/A N/A READ


The following table lists the access authority that PowerExchange CDC jobs and started tasks require for data sets that are dynamically allocated to system-generated DD names:

DTERIOME* N/A UPDATE N/A N/A

DTERIOMO* N/A UPDATE N/A N/A

DTERHLPE* N/A UPDATE N/A N/A

DTERHLPO* N/A UPDATE N/A N/A

DTLAMCPR UPDATE N/A N/A READ

DTLCACDC UPDATE N/A N/A UPDATE

DTLCACDE UPDATE N/A N/A READ

DTLCACFG N/A N/A N/A READ

DTLCAMAP UPDATE N/A N/A READ

DTLCFG READ READ N/A READ

DTLKEY READ READ N/A READ

DTLLOG UPDATE UPDATE N/A UPDATE

DTLLOGnn UPDATE UPDATE N/A UPDATE

DTLMSG READ READ N/A READ

DTLSGN READ N/A N/A READ

EDMMSG* UPDATE N/A UPDATE UPDATE

EDMPARMS READ READ READ READ

EDMSCTL N/A READ N/A N/A

EDMSLOG* N/A UPDATE N/A N/A

ERDS01 N/A N/A UPDATE N/A

ERDS02 N/A N/A UPDATE N/A

* Dynamically allocated SYSOUT data set.

Job or Started Task

Data Sets Authority Required

PowerExchange Agent

CCT data set, which contains capture registrations UPDATE

PowerExchange Logger

PRILOG and SECLOG log data sets UPDATE

Archive log data sets ALTER

PowerExchange Condense

CDCT data set UPDATE

Checkpoint data sets ALTER

Condense files ALTER

IMS log-based ECCR

RECON data sets processed by the IMS log-based ECCR READ

IMS secondary log data sets (SLDS) and IMS DLI batch log data sets (//IEFRDR) contained in the RECON data sets processed by the IMS log-based ECCR.

READ


PowerExchange Agent Authority

PowerExchange Logger Authority


MVS Security 93

Data Set Access Requirements for PowerExchange ECCRs

You must authorize access to the data sets used by PowerExchange ECCRs to the user ID under which the ECCR run. PowerExchange ECCRs allocate data sets by using DD statements in JCL, and also dynamically allocate data sets during execution.

The following table lists the authority that the Adabas, DB2, and IMS ECCRs require for data sets, by DD name:

The following table lists the access authority that the batch VSAM, CICS/VSAM, and Datacom synchronous ECCRs require for data sets, by DD name:

DD NameAdabas ECCR Authority

Adabas User Exit 2 Job Authority

DB2 ECCR Authority

IMS Log-Based ECCR Authority

IMS Synchronous ECCR Authority

STEPLIB READ READ READ READ READ

DATAMAP READ N/A N/A READ N/A

DDASSOR1 UPDATE N/A N/A N/A N/A

DDCARD READ N/A N/A N/A N/A

DDDATAR1 UPDATE N/A N/A N/A N/A

DDWORKR1 UPDATE N/A N/A N/A N/A

DTLADKSD READ N/A N/A N/A N/A

DTLAMCPR READ N/A N/A READ N/A

DTLCACDC N/A N/A N/A READ N/A

DTLCACFG READ N/A N/A READ N/A

DTLCACHG N/A N/A N/A READ N/A

DTLCCADA N/A UPDATE N/A N/A N/A

DTLCCPLG N/A READ N/A N/A N/A

DTLCFG READ READ N/A READ N/A

DTLKEY READ READ N/A READ N/A

DTLLOG UPDATE UPDATE N/A UPDATE N/A

DTLMSG READ READ N/A READ N/A

EDMMSG* UPDATE N/A UPDATE UPDATE UPDATE

EDMPARMS READ N/A READ READ READ

REPL2CTL N/A N/A READ N/A N/A

REPL2OPT N/A N/A READ N/A N/A

* Dynamically allocated SYSOUT data set

DD NameBatch VSAM ECCR Authority

CICS/VSAM ECCR Authority

Datacom Change Collector Authority

Datacom Change Controller Authority

Datacom Log Feeder Authority

STEPLIB READ READ READ READ READ

DFHRPL N/A READ N/A N/A READ

DTLAMCPR N/A N/A N/A READ N/A

DTLCACDC N/A N/A N/A READ N/A

DTLCACFG N/A N/A N/A READ N/A


The following table lists the access authority that the Datacom table-based and IDMS log-based ECCRs require for data sets, by DD name:

Security for Capture Registrations and Extraction Maps PowerExchange checks PowerExchange-specific resource profiles to determine whether to permit access to capture registrations and extraction maps. For this security checking to occur, you must specify 2 in the first parameter of SECURITY statement in the DBMOVER configuration file. PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.

By default, PowerExchange checks resource profiles in the FACILITY class. Use the RACF_CLASS parameter in the DBMOVER configuration file to specify a different class for checking security profiles. If you specify a different class, it must be defined in RACF or the security product you use, with a resource name that has a maximum length of 128 bytes.

Restriction: If you specify 2 in the first parameter of SECURITY statement, you must specify resource profiles for capture registrations and extraction maps. Otherwise, PowerExchange fails access to these resources.

Resources Profiles for Capture Registrations and Extraction Maps

Configure resource profiles for capture registrations and extraction maps and set the access authorities.

For capture registration, a resource profile has the following form:

CAPX.REG.dbtype.dbid.registration_name

DTLCACHG N/A N/A N/A READ N/A

DTLCFG N/A N/A N/A READ N/A

DTLLOG N/A N/A N/A UPDATE N/A

DTLMSG N/A N/A N/A READ N/A

EDMMSG* UPDATE UPDATE UPDATE UPDATE UPDATE

EDMPARMS READ READ N/A N/A READ


DD NameDatacom Table-Based

IDMS Log-Based

STEPLIB READ READ

DTLAMCPR READ READ

DTLCACFG READ READ

DTLCFG N/A READ

DTLLOG UPDATE UPDATE

DTLLOGnn UPDATE UPDATE

DTLKEY READ READ

DTLMSG READ READ

EDMMSG* UPDATE UPDATE

EDMPARMS READ READ


DD NameBatch VSAM ECCR Authority

CICS/VSAM ECCR Authority

Datacom Change Collector Authority

Datacom Change Controller Authority

Datacom Log Feeder Authority

MVS Security 95

For extraction maps, a resource profile has the following form:

CAPX.CND.dbid.registration_name.table_name

The following table describes the variables in the resource profiles:

To access capture registrations and extraction maps, use the following access authorities in resource profiles:

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers all capture registration and extraction map profiles:

CAPX.**

Security for Data Maps PowerExchange checks PowerExchange-specific resource profiles to determine whether to permit access to data maps. For this security checking to occur, you must specify the following parameters in the DBMOVER configuration file:

♦ 2 in the first parameter of SECURITY statement

♦ Y for the DM_SUBTASK parameter

PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.


Parameter Valid Values Description

dbtype ADA Adabas

DB2 DB2

DCM Datacom

IDM IDMS

IMS IMS

VSM VSAM

dbid Character string Instance, Database ID, or Collection ID that you specified in the registration group that contains the capture registration or the extraction group that contains the extraction map.

registration_ name Character string Capture registration name.

table_ name Character string For DB2, table name for the registered table. Otherwise, table name from the data map for the registered source.

Access Type Description

READ Allows users to display capture registrations or extraction maps.

For extraction maps, allows users to extract change data.

UPDATE For capture registrations, allows users to edit capture registrations by changing the columns selected, the registration status, or the condense type.

For extraction maps, allows users to define new extraction maps, and delete or edit existing extraction maps.

ALTER Allows users to define, delete, and edit capture registrations.


Restriction: If you specify 2 in the first parameter of SECURITY statement and Y for the DM_SUBTASK parameter, you must specify a resource profile for data maps. Otherwise, PowerExchange fails access.

Resource Profiles for Data Maps

Configure resource profiles for data maps and set the access authorities.

The resource profile for access to data maps has the following form:

DTL.DATAMAP.resource_name

The resource_name variable specifies the suffix for the resource profile. This suffix is the value specified for the DM_RESOURCE parameter in the DBMOVER configuration file, or if DM_RESOURCE is not specified, the default of DATASET.

To control access to data maps, use the following access authorities:

If you do not define any resource profiles for data maps, PowerExchange denies access.

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers all DM_RESOURCE values:

DTL.DATAMAP.**

Security for PowerExchange Listener CommandsPowerExchange checks PowerExchange-specific resource profiles to determine whether to permit access to PowerExchange Listener commands issued from the PowerExchange Navigator or the DTLUTSK utility. For this security checking to occur, you must specify 2 in the first parameter of SECURITY statement in the DBMOVER configuration file. PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.


Restriction: If you specify 2 in the first parameter of SECURITY statement, you must specify a resource profiles for PowerExchange Listener commands. Otherwise, PowerExchange fails PowerExchange Listener commands issued from the PowerExchange Navigator or the DTLUTSK utility.

Resources Profiles for PowerExchange Listener Commands

Configure resource profiles for PowerExchange Listener commands and set the access authorities.

The resource profile for the LISTTASK command has the following form:

DTL.TASKCTRL.DISPLAY

The resource profile for the STOPTASK command has the following form:

DTL.TASKCTRL.STOPTASK

Users that have READ access to the resource profiles can use the PowerExchange Navigator or the DTLUTSK utility to issue LISTTASK or STOPTASK commands. Otherwise, access to these commands is denied. If you do not define any resource profiles, PowerExchange denies access to these commands.

Access Type Description

READ Allows users to read data maps

UPDATE Allows users to define, delete, and modify data maps

MVS Security 97

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers all PowerExchange Listener commands that you can issue from the PowerExchange Navigator or the DTLUTSK utility:

DTL.TASKCTRL.**

Adabas SecurityYou can use the following Adabas security options with PowerExchange:

♦ Password security.

♦ SAF security

♦ Security by Value

To use Adabas password security, specify file passwords in PowerExchange data maps. To use Adabas SAF security or security by value, no action is required in PowerExchange.

Restriction: PowerExchange does not support Natural Security, which is a Natural application. PowerExchange supports Adabas data access directly, not through an application layer.

PowerExchange checks PowerExchange-specific resource profiles to determine whether to permit write access to Adabas files. For this security checking to occur, you must specify 2 in the first parameter of SECURITY statement in the DBMOVER configuration file. PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.


Resource Profiles for Controlling Write Access to Adabas Files

Configure resource profiles for write access to Adabas files and set the access authorities.

The resource profile for controlling write access to Adabas files has the following form:

DTL.DBWRITE.ADABAS.DBdbid.FNfile_num

where:

♦ dbid is a physical database ID for the database that contains the file. This value begins with “DB” and is followed by the database ID, which must be left-padded with zeroes so that the ID has five digits. Valid values are 00001 through 99999.

♦ file_num is a file number. This value begins with “FN” and is followed by the file number, which must be left-padded with zeroes so that the file number has five digits. Valid values are 00001 through 99999.

Users that have UPDATE access to the resource profile can use PowerExchange to write data to the Adabas file. Otherwise, access is denied. If you do not define any resource profiles, PowerExchange does not control write access.

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers all Adabas files in all databases:

DTL.DBWRITE.ADABAS.**

The following generic profile in RACF covers all Adabas files in databases 1,000 through 1,999:

DTL.DBWRITE.ADABAS.DB01*.**


Datacom SecurityPowerExchange checks PowerExchange-specific resource profiles to determine whether to permit read access to Datacom tables. For this security checking to occur, you must specify 2 in the first parameter of SECURITY statement in the DBMOVER configuration file. PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.


Resource Profiles for Controlling Read Access to Datacom Tables

Configure resource profiles for read access to Datacom table and set the access authorities.

The resource profile for controlling read access to Datacom tables has the following form:

DTL.DBREAD.DATACOM.Ddatabase_id.short_table_name

where:

♦ database_id is a database identifier (ID) that begins with “D” and is followed by five digits. Database IDs must be left-padded with zeroes so that the ID always has five digits. PowerExchange matches the database ID in the resource profile against the database IDs of the actual Datacom databases. Valid values are 00001 through 99999.

♦ short_table_name is the short table name that is defined in Datacom for a table. PowerExchange matches the specified short table name against Datacom short table names and indirectly against Datacom long table names to determine the tables for which RACF is to restrict READ access. You do not need to define separate RACF resource profiles for long table names. If you create resource profiles for long table names only, PowerExchange cannot also match on short table names to identify the tables that RACF is to protect.

Users that have READ access to the resource profile can use PowerExchange to read data from the Datacom table or obtain metadata for the table. Otherwise, access is denied. If you do not define a resource profile, PowerExchange does not control read access. For example, if you have a database ID of 23 and short table name of STL, create the following resource profile:

DTL.DBREAD.DATACOM.D00023.STL

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers all Datacom tables in all databases:

DTL.DBREAD.DATACOM.**

If an unauthorized user attempts a task such as a database row test of a data map or extraction map in PowerExchange Navigator, PowerExchange issues the following message:

PWX-00243 Userid user_id not authorized to access resourceDTL.DBREAD.DATACOM.Ddatabase_id.short_table_name req=R by RACF.

In this message, the req value of “R” represents READ access.

If an unauthorized user attempts a task such as creating a capture registration or data map, PowerExchange issues the following message:

PWX-02530 The metadata requested was not available, please check your parameters.

DB2 for z/OS SecurityTo use connection user IDs control access to DB2 tables and resources, you must specify the following statement and parameter in the DBMOVER configuration file on the MVS system:

♦ 2 in first parameter of the SECURITY statement

♦ RRSAF for the MVSDB2AF parameter

MVS Security 99

Otherwise, PowerExchange uses the user ID under which the PowerExchange Listener or Netport job runs to control access to DB2 tables and resources.

Note: To specify RRSAF for the MVSDB2AF parameter, the Resource Recovery Services (RRS) component of MVS must be active.

IMS SecurityPowerExchange checks PowerExchange-specific resource profiles to determine whether to permit write access to IMS databases. For this security checking to occur, you must specify 2 in the first parameter of SECURITY statement in the DBMOVER configuration file. PowerExchange then uses the MVS System Authorization Facility (SAF) interface to check access to the resource profiles.


Resource Profiles for Controlling Write Access to IMS Databases

Configure resource profiles for write access to IMS databases and set the access authorities.

The resource profile for controlling write access to IMS databases access through DL/I batch has the following form:

DTL.DBWRITE.IMS

The resource profile for controlling write access to IMS databases access through ODBA has the following form:

DTL.DBWRITE.IMS.ims_id

The ims_id variable specifies the IMS SSID value that you specify on the Data Maps Properties dialog box in the PowerExchange Navigator. Although you can also specify an IMS SSID value for DL/1 batch through the PowerExchange Navigator, this value is not used in the resource profile specification for DL/1 batch. For more information about specifying data map properties, see the PowerExchange Navigator User Guide.

Users that have UPDATE access to the resource profile can use PowerExchange to write data to IMS databases. Otherwise, access is denied. If you do not define a resource profile, PowerExchange does not control write access.

You can define generic resource profiles to minimize the number of definitions. For example, the following generic profile in RACF covers writing to IMS databases with DL/I batch or ODBA:

DTL.DBWRITE.IMS.**

You can use this type of generic profile to prevent general write access.

PowerExchange Selective Sign-on

You can use PowerExchange selective sign-on to control connection access to PowerExchange. With selective sign-on, PowerExchange checks the PowerExchange sign-on file to verify access for the supplied user ID and, optionally, the IP address. PowerExchange performs selective sign-on checking when a PowerExchange Listener accepts a new connection over TCP/IP, and prior to starting a subtask for the request.

Tip: On MVS and i5/OS, PowerExchange can also utilize security facilities of the MVS and i5/OS operating systems to authenticate users. Use PowerExchange selective sign-on in conjunction with the user authentication option of the SECURITY statement. Selective sign-on does not perform password validation and provides only a minimal level of security.


To configure selective sign-on for a PowerExchange Listener:

1. Specify Y for the second parameter of the SECURITY statement in the DBMOVER configuration file.

Note: The value of the first parameter of the SECURITY statement does not affect whether PowerExchange uses the PowerExchange sign-on file to control connection access to PowerExchange or access to PowerExchange Listener commands issued through the PowerExchange Navigator. However, it does control whether PowerExchange uses the sign-on file to authorize use of pwxcmd commands.

2. Set up a PowerExchange sign-on file on the system on which the PowerExchange Listener runs. In the file, specify the allowed users and, optionally, the IP addresses from which those users are allowed access. For more information, see “PowerExchange Sign-on File” on page 101.

3. Start or restart the PowerExchange Listener.

PowerExchange Sign-on File

The PowerExchange sign-on file authorizes users to perform the following tasks:

♦ Connect to PowerExchange systems from specific IP addresses

♦ On Linux, UNIX, and Windows, issue specific pwxcmd commands

♦ On Windows, issue PowerExchange Listener commands to a PowerExchange Listener through the PowerExchange Navigator

PowerExchange Sign-on File SyntaxUse the following syntax for the sign-on file statements:

[AUTHGROUP=(group_id,COMMANDS=(command1,command2,...))]USER=(user_ID[@domain] ALLOW={F|N|Y} [AUTHGROUP=group_id] [IP=nnnn.nnnn.nnnn.nnnn] [TASKCNTRL={D|S}] )

The following rules and guidelines apply:

♦ The maximum statement length is 64 characters, and can begin in any column.

♦ Statements cannot be continued and must be coded on a single line.

♦ You can specify one or more AUTHGROUP or USER statements in a sign-on file.

♦ When PowerExchange encounters a USER statement that matches all or part of the user ID, PowerExchange stops reading the sign-on file.

♦ Comment lines must begin with /*, and must be on a separate line from any statements.

PowerExchange provides a sample sign-on file, in the required location, that you can customize. The following table describes, by operating system, the name and required location of sign-on file:

Operating System

SIGNON File Name and Location

i5/OS SIGNON member in the datalib /CFG file

Linux, UNIX, and Windows

signon.txt file in the PowerExchange installation directory

MVS SIGNON member in RUNLIB to which the DTLSGN DD in the PowerExchange Listener JCL points

PowerExchange Sign-on File 101

PowerExchange Sign-on File StatementsConfigure the following statements to authorize selective sign-on, PowerExchange Listener commands through PowerExchange Navigator, or pwxcmd commands:

♦ For selective sign-on and PowerExchange Listener command authorization, use the USER statement.

♦ For pwxcmd command authorization on Linux, UNIX, and Windows, use the USER and AUTHGROUP statements.

AUTHGROUP Statement

Optionally, use the AUTHGROUP to limit access to specified pwxcmd commands to the users in the authorization group.

The AUTHGROUP statement has the following syntax:

AUTHGROUP=(group_id,COMMANDS=(command1,command2,...))

The AUTHGROUP statement has the following parameters:

group_idRequired. Specifies the name of the authorization group. When you define USER parameters to assign users to this authorization group, include the AUTHGROUP parameter with this same group ID.

COMMANDS=(command1,command2,...)Required. Specifies a list of pwxcmd commands that users in this authorization group are authorized to run.

For the PowerExchange Listener, specify one or more of the following commands:

♦ CLOSE

♦ CLOSEFORCE

♦ LISTTASK

♦ STOPTASK

For the PowerExchange Logger for Linux, UNIX, and Windows, specify one or more of the following commands:

♦ CONDENSE

♦ DISPLAYSTATUS

♦ FILESWITCH

♦ SHUTCOND

♦ SHUTDOWN

To authorize users to run all of these commands, specify asterisk (*) instead of commands.

Note: You do not need authorization to run the pwxcmd HELP and VERSION commands.

USER Statement

Use the USER statement to authorize the specified user and, optionally, IP address, to connect to PowerExchange. Optionally, on Linux, UNIX, and Windows, use the USER statement to authorize the specified user to issue pwxcmd commands. On Windows, use the USER statement authorize users to issue PowerExchange Listener commands through the PowerExchange Navigator.The maximum statement length is 64 characters.

The USER statement has the following syntax:

USER=(user_ID[@domain] ALLOW={F|N|Y} [AUTHGROUP=group_id] [IP=nnnn.nnnn.nnnn.nnnn] [TASKCNTRL={D|S}] )


The USER statement has the following parameters:

user_id[@domain]Required. Specifies the user ID of a user. Optionally, on Windows, you can specify a domain name. User ID checking is case-sensitive.

You can use the following wildcard characters in the user ID:

♦ Asterisk (*), which represents 0 or more characters.

♦ Question mark (?), which represents exactly one character.

If the user ID contains an asterisk (*) or question mark (?) character as a part of the ID, precede these characters with the tilde (~) escape character.

The user specifications on a pwxcmd command and the USER statement in the sign-on file must match. If you specify a domain as part of a user ID in the sign-on file, you must also specify the domain in the user ID when you issue a pwxcmd command.

If you do not specify a domain as part of a user ID in a pwxcmd command, PowerExchange assumes that the user ID is local or a member of the same domain as the user ID under which the PowerExchange Listener runs. If you run the PowerExchange Listener as a Windows service and that service runs under the local user, you must specify the domain as part of a user ID in a pwxcmd command.

ALLOW={F|N|Y}Required. Specifies one of the following values:

♦ F. Indicates that additional ALLOW parameters follow.

♦ N. The user cannot connect on to the PowerExchange. If an IP address is specified, the user is restricted from that IP address only.

♦ Y. The user can connect to PowerExchange from any IP address, unless a specific IP address is specified in the IP parameter.

AUTHGROUP=group_IDOptional. If you are controlling access to pwxcmd commands, specifies the name of the authorization group to which this user is assigned. This name must also be specified in the AUTHGROUP statement.

IP=nnnn.nnnn.nnnn.nnnnOptional. Specifies an IP address of a user. The nnnn value of an IP address, called an octet, is a number from 0 through 255.

If the last octet of the IP address is zero, then PowerExchange use the octet values to cover number from 0 to 255. For example, the following IP parameter specification covers all IP addresses from 10.10.10.0 to 10.10.10.255:

IP=10.10.10.0

If an octet begins with zero, then PowerExchange assumes the value specified is octal. For example, if you specify 0120, PowerExchange coverts this value to 80. Generally, you should specify decimal numbers for the octet values.

TASKCNTRL={D|S}Optional. On Windows, authorizes the user to issue the LISTTASK and STOPTASK commands from the PowerExchange Navigator. Valid values are:

♦ D. The user can issue the LISTTASK command from the PowerExchange Navigator.

♦ S. The user can issue the STOPTASK command from the PowerExchange Navigator.

This option is valid only if SECURITY=(1,Y) in the DBMOVER configuration file.

To issue a command from the PowerExchange Navigator, click TASK_CNTL in the DB Type list in the Database Row Test dialog box. For more information, see thePowerExchange Navigator User Guide.

PowerExchange Sign-on File 103

PowerExchange Sign-on File ExamplesThe following topics show how to code the PowerExchange sign-on file to control connection to PowerExchange and access to pwxcmd commands.

PowerExchange Selective Sign-on Examples

The following example USER statements in a sign-on file show how to control connection to PowerExchange based on user ID and IP address.

To allow all users to connect to PowerExchange, code the following USER statement in the sign-on file:

USER=(* ALLOW=Y)

To restrict access to PowerExchange to users with user IDs that begin with SYS, code the following USER statements in the sign-on file:

USER=(SYS* ALLOW=Y)USER=(* ALLOW=N)

To allow users to sign on from multiple IP addresses, specifying multiple USER statements with the ALLOW=F option. When PowerExchange encounters the first USER parameter with ALLOW=Y or N and a user ID that matches, PowerExchange stops reading the sign-on file. The following example USER statements allow USER1 to connect to PowerExchange from three different IP addresses:

USER=(USER1 ALLOW=F IP=10.7.16.25) USER=(USER1 ALLOW=F IP=10.7.16.26) USER=(USER1 ALLOW=Y IP=10.7.16.30)

In the following example, the second USER statement specifies ALLOW=Y. As a result, USER1 can sign on only from two IP addresses: 10.7.16.25 and 10.7.16.26.

USER=(USER1 ALLOW=F IP=10.7.16.25) USER=(USER1 ALLOW=Y IP=10.7.16.26) USER=(USER1 ALLOW=Y IP=10.7.16.30)

pwxcmd Command Authorization Examples

On Linux, UNIX, and Windows, you can use the sign-on file to control which users can run specific pwxcmd commands. In the sign-on file, define AUTHGROUP and USER statements to authorize users to run pwxcmd commands. The AUTHGROUP statement specifies the authorization group and which pwxcmd commands to control. The USER statement associates a user ID with an authorization group and specifies whether to allow the user to run the specified commands.

The following example assigns user ops99 to the ADMIN authorization group, enables this user to connect to PowerExchange, and to run all pwxcmd commands:

AUTHGROUP=(ADMIN, COMMANDS=(*))USER=(ops99 ALLOW=Y AUTHGROUP=ADMIN)

The following example assigns users with user IDs that begin with SYS to the OPER authorization group, which authorizes these users to run the specified pwxcmd commands, and allows these users to connect to PowerExchange:

AUTHGROUP=(OPER, COMMANDS=(close, closeforce, listtask, stoptask))USER=(SYS* ALLOW=Y AUTHGROUP=OPER)

The following example allows all users to connect to PowerExchange and to run the pwxcmd LISTTASK command:

AUTHGROUP=(BASIC, COMMANDS=(LISTTASK))USER=(* ALLOW=Y AUTHGROUP=BASIC)


C H A P T E R 7

Secure Sockets Layer Support


♦ SSL Security on a PowerExchange Network, 105

♦ Configuring SSL, 107

♦ Sample Certificate Authority List (CALIST), 109

♦ AT-TLS Policy File for PowerExchange, 109

SSL Security on a PowerExchange Network

You can configure Secure Sockets Layer (SSL) communication on each machine in the PowerExchange network to ensure secure communication. You can configure SSL communication by establishing certificates and keys that authorize the secure connection between machines and enable encryption and decryption of data.

The client authorizes the servers to use SSL communication by means of certificates that are saved in its Certificate Authority list. The Certificate Authority list contains the public key certificates for any server that it wishes to communicate with through SSL. The following keys are used:

♦ Public key certificates. Public key certificates, held in the Certificate Authority List for a machine, are distributed to all other machines to authenticate the identity of the machine.

You can use the public key to encrypt the initial handshake message.

♦ Private keys. Machines on the secure network use private keys to decrypt handshake messages sent to the machine.

SSL Security on a PowerExchange Network 105

Figure 7-1 shows how SSL can be configured in a PowerExchange environment:

Each server or client machine has SSL private key and SSL certificate components. You can activate these components by configuring PowerExchange. For more information about configuring PowerExchange, see “Configuring SSL” on page 107.

After configuration, the initial handshake and acceptance set up the secure connection. The individual data messages are encrypted using a unique key that is generated by the sender. The receiver is able to decrypt the data message.

Application Transparent - Transport Layer Security (AT-TLS)On z/OS release 1.7 and later, AT-TLS uses a Communications Server policy file to determine which sessions will use the SSL protocol. The policy is based on attributes such as server job name, IP addresses, and the SSL attributes to be used for the sessions. This could be, for example, security parameters.

If AT-TLS is not in use for other applications, set up the AT-TLS infrastructure.

To set up the AT-TLS infrastructure:

1. Activate the Communications Server Policy Agent.

2. Create the AT-TLS policy file for PowerExchange and make it available to the Policy Agent.

For more information about AT-TLS, see the IBM AT-TLS documentation.

IBM provides the Network Configuration Assistant tool to assist in building an AT-TLS configuration and policy files. This is a graphical user interface which can be downloaded from the IBM z/OS support web site. A sample policy file is provided in “AT-TLS Policy File for PowerExchange” on page 109.

Obtain SSL certificates for all machines that use SSL connection in the PowerExchange environment. The system administrator will supply these certificates.

Figure 7-1. Example Setup on PowerExchange Client and Server Platforms

PowerExchangeClient

Configuration filedbmover.cfg

PowerExchangeClient

Windows, UNIX or Linux

PowerExchangeServer

Configuration filedbmover.cfg

Windows, UNIX or Linux

PowerExchangeServer

ListenerSSL Mode

CertificateAuthority List

SSLCertificate

MVS

AT-TLS

PowerExchangeServer

ListenerSSL Mode


SSLCertificate


SSLCertificate

Private KeyDecrypt

Private KeyDecrypt

Private KeyDecrypt

SSL Component

ServerEnvironment

PowerExchangeComponent

ClientEnvironment

Key

106 Chapter 7: Secure Sockets Layer Support

Configuring SSL

To configure SSL on the PowerExchange platform, you must add the PowerExchange Listener and Node lines, and the SSL parameter, to the DBMOVER configuration file on the client and server machines.

Note: Use the DTLUTSK utility, or a row test in the PowerExchange Navigator, to view the SSL status of a PowerExchange Listener. For more information about the DTLUTSK utility, see the PowerExchange Utilities Guide. For more information about Row Test, see the PowerExchange Navigator User Guide.

Configuring SSL Client and Server on MVSThe MVS client and server do not require configuration file changes. PowerExchange on MVS uses AT-TLS to implement SSL. For more information about AT-TLS, see the AT-TLS product documentation.

Configuring SSL Server on Windows, UNIX, or LinuxThe LISTENER line specifies the parameters for the PowerExchange Listener which is operating in SSL mode:

LISTENER=(<node>,TCPIP,<port_number>,,,,,,,SSL)

Note: To avoid command failure, maintain the relative position of the SSL parameter. Note that there are six empty parameters between the port number and SSL.

The SSL parameter specifies the SSL certificate that you are using to make the SSL connection:

SSL=(KEY=server.pem)

For more information about the LISTENER parameter, see “LISTENER” on page 20.

Configuring the SSL Client on Windows, UNIX, or LinuxThe NODE line specifies the server you want to connect to in SSL mode.

NODE=(<server_listener>,TCPIP,<remote_host>,<port_number>,,,,,,SSL)

or

NODE=(<server_listener>,TCPIP,<remote_host>,<port_number>,,,,,,ZOSSSL)

Use the ZOSSSL parameter when accessing an MVS platform. Use the SSL parameter to access all other platforms. Use the SSL parameter, rather than the ZOSSSL parameter, if PTFs UK26131 (z/OS 1.8) or UK26132 (z/OS 1.9) have been installed on the MVS machine to rectify APAR PK46403. For more information about APAR PK46403, see the IBM web site.

Note: To avoid command failure, maintain the relative position of the SSL or ZOSSSL parameter. Note that there are five empty parameters between the port number and SSL or ZOSSSL.

For more information about the NODE parameter, see “NODE” on page 24.

The SSL parameter specifies the SSL key, pass phrase and Certificate Authority list (CALIST) that you are using to make the SSL connection. For example:

SSL=(PASS=client,KEY=client.pem,CALIST=root.pem)

CALIST can be replaced with CAPATH. For example:

SSL=(PASS=client,KEY=client.pem,CAPATH=/pwx/certs/)

To optimize performance, use CAPATH to specify the location of multiple certificates.

Use CAPATH to specify the trusted CA directory of the OpenSSL installation on Linux or UNIX.

Use CAPATH to specify the ‘certs’ directory of the OpenSSL installation on Windows.

Configuring SSL 107

Running Secure and Non-secure ModesYou can run a PowerExchange Listener in SSL mode and a PowerExchange Listener in non-SSL mode by specifying the PowerExchange Listener NODE twice in the Client configuration file. For example:

NODE=(MVS2,TCPIP,10.7.16.35,9764,,,,,,SSL)NODE=(MVS2,TCPIP,10.7.16.35,9765)

The first PowerExchange Listener will monitor SSL communication on port 9764. The second PowerExchange Listener will monitor non-SSL communication on port 9765.

SSL Authentication ModeThe alternative modes of Certificate Authentication that can be specified are as follows:

♦ Request client authentication. By default, PowerExchange uses client authentication. This means that the server authenticates the identity of the client. The server requests the client certificate and checks that it can be validated against its certificate authority list. For example:

− On the server:

SSL_REQ_CLNT_CERT=Y

− On the client:

SSL_REQ_SRVR_CERT=N

If these parameters are not specified, PowerExchange uses client authentication.

♦ Request server authentication. Request Server Authentication means that the client checks the identity of the server. To enable Request Server Authentication you must set the SSL Authorization mode configuration parameters in the DBMOVER configuration file. For example:

− On the server:

SSL_REQ_CLNT_CERT=N

− On the client:

SSL_REQ_SRVR_CERT=Y

♦ No authentication. Neither Client or Server requests certificates or checks that the certificates can be validated against its certificate authority list. SSL protection is still in place. To enable this form of authentication you must set the SSL Authorization mode configuration parameters. For example:

− On the server:

SSL_REQ_CLNT_CERT=N

− On the client:

SSL_REQ_SRVR_CERT=N

♦ Client and server authentication. Client and Server Authentication means that the client checks the identity of the server, and that the server checks the identity of the client. To enable this form of authentication you must set the SSL Authorization mode configuration parameters. For example:

− On the server:

SSL_REQ_CLNT_CERT=Y

− On the client:

SSL_REQ_SRVR_CERT=Y


Sample Certificate Authority List (CALIST)

An installation for a particular machine may have a private key and a Certificate Authority list. The Certificate Authority list comprises number of certificates giving access to other platforms. The keys and certificates are fully encrypted.

Private Key-----BEGIN RSA PRIVATE KEY-----Proc-Type: 4,ENCRYPTEDDEK-Info: DES-EDE3-CBC,0490DA0C0B08D5EF

7x5PehkSQy0MeQR/SbyTKcY4J78SagF9qqVzGPQVguVm3r+ubJDITLm1W9ae8vQ/Hn4QBC86+tOKZLHoZ4BpkNG4eRawcqm3zBjWVWVbh81+xqLxB6mCvmqRauIMxjPIxc7l2vJidv/J4LMoYmG1UvZPZ8YtD7QDDzM+f1EHlNOKXe139pLdLRJXuE5mdoXcD8+EglUm226yV99ZbtsZJrNJAJI7AxlVwu2R7KWsNPz2S8cz1e7Vp/jTVaFS5hemkoTHEquhnIwDbsnyCIEu/ynAK3nNd1tKKKQnPos+T/Z2EMcAfXTRnXIXPk2mRvcHuWFujBeMtiySJ73qEsyG+7fk3pxo5Mknp/9Bejc3Fn8sJdGpVckMuW9fZLWr7h++fmD0RN/zD0fatFlfsM0geUtsBHLqWixFzA0ikaKvA1m3Q4ZaVhEZwIZ0So1fN5+2-----END RSA PRIVATE KEY-----

Certificate Authority List-----BEGIN CERTIFICATE-----MIIC5DCCAcwCAQAwDQYJKoZIhvcNAQEFBQAwODELMAkGA1UEBhMCVUsxEzARBgNVBAgTClNvbWUtU3RhdGUxFDASBgNVBAoTC0luZm9ybWF0aWNhMB4XDTA3MDIwODE0MjgwOFoXDTA3MDIxMDE0MjgwOFowODELMAkGA1UEBhMCVUsxEzARBgNVBAgTClNvbWUtU3RhdGUxFDASBgNVBAoTC0luZm9ybWF0aWNhMIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAtJ0qYgZ3ofhI3m8iNtLSoEhyzzyUVmp7KTqRLfrAegED3+dj+pYL8RXj1xvKQItCVY9xwkoXKqQI9rNfSnSJhZr0XTvMx6Lj5V5s5ZG2OVZNf/xSNPXr8oObBTszqtPvsjkq/+eskFIvLxgCvOqz5z+odizqvaRsUwWqr2HW9Pt9-----END CERTIFICATE----------BEGIN CERTIFICATE-----MIICvzCCAiigAwIBAgIBADANBgkqhkiG9w0BAQUFADBiMQswCQYDVQQGEwJHQjEUMBIGA1UEChMLSW5mb3JtYXRpY2ExEjAQBgNVBAsTCXpPUy5BZG1pbjEpMCcGA1UEAxMgaXJyY2VydGEubWh6ODkwMi5pbmZvcm1hdGljYS5jb20wHhcNMDYxMTA2MDAwMDAwWhcNNDAxMjMxMjM1OTU5WjBiMQswCQYDVQQGEwJHQjEUMBIGA1UEChMLSW5mb3JtYXRpY2ExEjAQBgNVBAsTCXpPUy5BZG1pbjEpMCcGA1UEAxMgaXJyY2VydGEubWh6ODkwMi5pbmZvcm1hdGljYS5jxojevZ8wDQYJKoZIhvcNAQEBBQADgY0AMIGJAoGBAMJ2SfrqOnMOLkYVHqRmDLCHR3IMfNHwH5oRIpN8Si+Tit6NhDcxjwm8wkEZ-----END CERTIFICATE-----

AT-TLS Policy File for PowerExchange

The policy file specifies the MVS jobs that are authorized to use SSL. Any job that communicates with an SSL Listener must also be authorized in the policy file. You can add multiple job names to the AT-TLS policy file.

The default location of the policy file is “/etc/pagent_TTLS.conf ”. For more information about AT-TLS policy files, see the IBM AT-TLS documentation.

The following is an example of a AT-TLS policy file for PowerExchange:

## ## AT-TLS Policy Agent Configuration file for: ## Image: MHZ2 ## Stack: TCPIP ## This is a sample AT_TLS Policy Agent Configuration file for Power Exchange, based ## on a subset of a file originally created by the IBM z/OS Network Security ## Configuration Assistant application. ## ## It defines a listener (jobname PWXSSL85) which acts as SSL server ## (Direction=inbound) and a second listener (PWXSSLMX) which acts as SSL server only ## when connected via its port 18501. ## ## TLSRule 0-1 defines listener with jobname PWXSSL85 which as as SSL server for all ## connections (connection direction is inbound - from client to listener). ## - IpAddrSet addr1 specifies that the rule applies to any IP address and port. ## - Group Action gAct1 specifies that SSL is to be enabled for this connection. ## - Environment Action eAct1 specifies that the listener acts as SSL server, with ## keyring defined by the keyR-MHZ2 TTLSkeyRingPArms statement.

Sample Certificate Authority List (CALIST) 109

## - Connection Action cAct1 specifies that encryption parameters defined by ## TTLSCipherParms statement cipher1~AT-TLS__Silver, and that the listener is to act ## as SSL server (without client certificates). ## ## TLSRule 0-5 defines a similar listener with jobname PWXSSLMX which has more than one ## listener port. It is to use SSL only if communicating via its port 18501 (portrange ## statement portR1. ## ## TTLSCipherParms cipher1~AT-TLS__Silver shows a list on Cipher Suites which can be used ## for the connections. ##TTLSRule 0~1 { LocalAddrSetRef addr1 RemoteAddrSetRef addr1 Jobname PWXSSL85 Direction Inbound Priority 255 TTLSGroupActionRef gAct1 TTLSEnvironmentActionRef eAct1 TTLSConnectionActionRef cAct1 }TTLSRule 0~5~ { LocalAddrSetRef addr1 RemoteAddrSetRef addr1 LocalPortRangeRef portR1 Jobname PWXSSLMX Direction Inbound Priority 251 TTLSGroupActionRef gAct1 TTLSEnvironmentActionRef eAct1 TTLSConnectionActionRef cAct1 }TTLSGroupAction gAct1{ TTLSEnabled On Trace 7 }TTLSEnvironmentAction eAct1 { HandshakeRole Server EnvironmentUserInstance 0 TTLSKeyringParmsRef keyR~MHZ2 }TTLSEnvironmentAction eAct2 { HandshakeRole Client EnvironmentUserInstance 0 TTLSKeyringParmsRef keyR~MHZ2 }TTLSConnectionAction cAct1 { HandshakeRole Server TTLSCipherParmsRef cipher1~AT-TLS__Silver Trace 7 }TTLSConnectionAction cAct2 { HandshakeRole Client TTLSCipherParmsRef cipher1~AT-TLS__Silver Trace 7 }TTLSKeyringParms keyR~MHZ2 { Keyring ATTLS_keyring }TTLSCipherParms cipher1~AT-TLS__Silver { V3CipherSuites TLS_RSA_WITH_DES_CBC_SHA V3CipherSuites TLS_RSA_WITH_3DES_EDE_CBC_SHA }IpAddrSet addr1 { Prefix 0.0.0.0/0 }PortRange portR1 { Port 18501 }


C H A P T E R 8

Encryption


♦ Modes of Encryption, 111

♦ Methods of Encrypting Data in PowerExchange, 112

Modes of Encryption

PowerExchange supports the following modes of data encryption when moving data:

♦ No encryption

♦ PowerExchange encryption

♦ DES encryption

♦ RC2 encryption

PowerExchangeThis is a PowerExchange proprietary encryption algorithm.

DESThe Digital Encryption Standard (DES) is a U.S. commercial encryption standard that has been available for over 15 years. The federal standard document FIPS PUB 46-2 describes this algorithm. The key consists of 56 random bits and 8 parity bits, forming a 64-bit, or 8-byte, key.

Triple DESTriple DES executes DES three times, which triples the number of bits in an encryption key. A number of different methods achieve this function. This technique is known as Encrypt-Decrypt-Encrypt (EDE). The decryption process in the middle stage of Triple DES encryption provides compatibility with DES. If the three keys are the same, the Triple DES operation is equivalent to a single DES encryption. That way, an application that has only DES capabilities can still communicate with applications that use Triple DES. If the three keys are different, the decryption in the middle will scramble the message further. It will not decrypt the first stage.

111

Triple DES decryption is the inverse operation of the previous sequence, that is, DES decryption followed by DES encryption and then another DES decryption.

RC2The RC2 cipher was developed by Ronald Rivest as an alternative to DES encryption. It is proprietary to RSA Security. The RC2 input key can be of any length from 1 to 128 bytes. The algorithm uses the input key to generate an effective key that is actually used for encryption purposes. Internally, the algorithm builds a key table based on the bits of the key data. The chosen number of effective key bits limits the number of possible key tables. The effective key size is variable and takes values from 1 bit up to 1024 bits.

The RC2 cipher is a variable-key-size block cipher. Though a DES key requires exactly eight bytes, an RC2 key can vary between one and 128 bytes. The larger the key, the greater the security. The RC2 cipher is called a block cipher because it encrypts 8-byte blocks.

Methods of Encrypting Data in PowerExchange

The following methods are available to use data encryption within PowerExchange.

♦ Set defaults in the DBMOVER configuration file.

♦ Use PowerExchange ODBC driver parameters.

Setting Defaults in the DBMOVER Configuration FileYou can set the encryption that PowerExchange will use as default by adding the ENCRYPT and ENCRYPTLEVEL parameters to the DBMOVER configuration file. The parameters must be included in the configuration file on the platform where the data movement process is to be executed.

For more information, see “DBMOVER Parameter Descriptions” on page 4.

Figure 8-1. 24-bytes Triple DES Key

112 Chapter 8: Encryption

C H A P T E R 9

Nonrelational SQL Syntax


♦ Overview, 113

♦ Supported Syntax, 113

♦ Metadata Syntax, 117

♦ DTLDESCRIBE Qualifiers, 129

Overview

To access nonrelational sources, such as an IMS database or VSAM file, use the PowerExchange Navigator to define various physical and logical records. PowerExchange accepts a SQL statement in the normal way from the client and passes it to the PowerExchange Listener. The PowerExchange Listener accesses the database and retrieves the rows. PowerExchange accepts only a subset of the extensive SQL syntax. You must use only the SQL that PowerExchange supports for the SQL requests to be accepted.

Supported Syntax

The basic syntax supported by nonrelational data sources is as follows:

select col_a, col_b, .. col_n from schema.mapname.tablename [ where condition ]

The schema.mapname is the data map name based on the data-map naming convention. The tablename is the user-defined name that is defined in the data map.

The WHERE clause is optional. The WHERE condition can be constructed of one or more clauses that comprise "simple comparisons" or other types of conditions. These clauses are joined using one or more of the Boolean operators AND, OR, and NOT. If parentheses do not indicate the order of evaluation, the default order of NOT > AND > OR is used.

A "simple comparison" can be one of the following:

♦ basic

♦ like

♦ between

♦ null

113

Basic Comparison A basic comparison is where a column is compared with either another column or a constant. Complex expressions such as (col1 + col2) or (col1 * 3) are not supported.

Examplecol1 = '123'

Valid Operators

♦ Character constants must be enclosed in single quotation marks.

♦ Binary constants must be enclosed in single quotation marks and preceded by “X,” for example, X'00'.

♦ Numeric constants can be specified with an optional leading sign and decimal point. Exponential floating point format is allowed.

Comparisons can be made only between operands of similar types:

♦ A numeric operand can be compared only with another numeric operand.

♦ A character operand can be compared with either a character or binary operand.

♦ A binary operand can be compared with either a character or binary operand.

♦ Date, time and timestamp (date-time) operands can be compared only with an identical column type or a character constant in the correct format.

Date-time constants must have the following formats:

♦ DATE YYYY-MM-DD

Single-digit DD and MM values are allowed. If the lower order portion of the date is omitted, it will default to the values specified as the Data Defaults within the Data Map used.

♦ TIME HH.MI.SS

Single-digit HH, MI, and SS values are also allowed. If the lower order portion of the time is omitted it will default: MI=00, SS=00.

♦ TIMESTAMP YYYY-MM-DD HH.MI.SS.NNNNNN

The NNNNNN value is the fractional seconds value. Single digits are allowed for DD, MM, HH, MI, SS and NNNNNN. The date and time values will default as defined above, with further defaults of HH=00, NNNNNN=000000.

You can enter only the date or only the time component with the other component defaulting. If insufficient information is available to deduce which component has been entered, it will be assumed to be the date component.

When character fields of different length are compared, the operands are made equal lengths by padding the shorter one with spaces, prior to the comparison being made.

Similarly for binary fields, when fields of different length are compared, the operands are made equal lengths by padding the shorter one with nulls (X'00'), prior to the comparison being made.

Comparisons involving character or binary columns are limited to columns less than 256 bytes in length.

Name Operator

= Equals

<> or != Not equal

> Greater than

>= Not less than

< Less than

<= Not greater than

114 Chapter 9: Nonrelational SQL Syntax

LIKE ComparisonThis comparison is a pattern-matching comparison. The comparison is true if the column name matches the pattern.

Example:

col1 [NOT] LIKE pattern

The pattern is a character constant that contains any combination of characters within which the percent sign (%) and the underscore character (_) have special significance:

♦ % represents a series of zero or more arbitrary characters. Entering %% is the same as entering %.

♦ _ represents exactly one arbitrary character.

♦ Any character other than % or _ represents a single occurrence of that character.

♦ Trailing blanks in the column or in the pattern will be ignored for comparison purposes.

For example::

Use an ESCAPE clause to define the escape character to be used to escape the “%” and “_“ characters in the LIKE statement so that they can be matched explicitly.

You can use ESCAPE '\' in order that a string with 'A%\_' would allow all values starting with 'A' and ending with '_' to be returned.

For example, to escape the patterns in the previous table:

BETWEEN ComparisonThis comparison is testing a column for being falling in side or outside (used with NOT) a range of values.

Example:

col1 [NOT] BETWEEN valueA AND valueB

In this example valueA and valueB could be another column or a constant of the same type as the column being compared against. It is equivalent to coding two simple conditions as follows:

[NOT] (col1 >= valueA AND col1 <= valueB)

NULL ComparisonThis comparison is testing a column for being NULL or NOT NULL. It is treated as though it is less than any other possible value therefore it cannot be equal to anything other than another NULL column.

Example:

col1 IS [NOT] NULL

Pattern %TAB% matches but not

TAB , ATAB, TABULAR and MY_TABLE MY_TAABLE

Pattern TA% matches but not

TAB or TABULAR ATAB or MY_TABLE

Pattern TA_LE matches only TABLE from the examples above

Pattern %\_TAB% matches but not

MY_TABLE MYTABLE

Supported Syntax 115

Allowed KeywordsExamples of allowable keywords are:

ALLDISTINCT

Accessing Data with Column Names the Same as SQL KeywordsYou can access data with column names the same as SQL keywords by using single quotes around the column name. For example:

SELECT ‘DISTINCT’ FROM TABLE

Using SQL with User-defined FieldsYou should not use the SQL keyword 'DISTINCT' if the data map includes user-defined fields. Attempting to use DISTINCT with user defined fields causes a message in the logfile 'SELECT DISTINCT not guaranteed with Expressions'.

If DISTINCT is essential to a query then the best approach is to create another table in the data map that contains only the fields that are actually required for that query.

IMS Call ConsiderationsThe potential to store data in a non-sequential order in HDAM, DEDB, and PHDAM databases causes complications when dealing with ranges of data. Ideally, if searching for values between valueA and valueB, the search starts at valueA and issues Get Next calls until valueB. This sort of processing is possible only in an indexed sequential format. Where the keys are not in sequential order, positioning on valueA might be beyond some candidate segments. In this case, the result set will be incomplete. The following example shows the problem:

Select * from IMSSC1.IMSMP1.IMSDEMOS Where KEY >= 'A' and KEY <= 'D'

In this example, the randomizer has placed the records physically in the following key sequence:

Positioning on the first value, in this case A, then issuing Get Next calls until the end value of D would return only A and D, with candidate value C physically residing in the data set before A, and B after D.

The same issue arises for SELECT statements such as:

Select * from IMSSC1.IMSMP1.IMSDEMOS Where KEY = 'D' or KEY = 'B' or KEY = 'A'

In this case, three GU requests are needed instead of one GN request with a range (>='B' and <='D'). In case of the range request, IMS positions on the 'B' segment and reads by means of the twin pointer forward and misses the 'C' and 'D' segment because they are physically stored before the 'B' segment. The number of GU requests depends on the number of predicates in the query.

Therefore, if a range is requested of a HDAM, DEDB or PHDAM database, PowerExchange is forced to process the complete database sequentially, selecting all the required records as they are found.

C A D E B H F


Metadata Syntax

PowerExchange has special SQL syntax, called "dtldescribe,” for accessing metadata. This syntax references "qualifiers." The meaning of these qualifiers depends on the source database.

DTLDESCRIBE TABLESEnables you to list the available tables:

dtldescribe tables, [qualifier1], [qualifier2], [tablename], [comments], [ORDER], [escape character], [respect case 1], [respect case 2], [respect case 3], [extended information], [access method]

After the basic syntax of dtldescribe tables, all the other parameters in brackets [ ] are optional. If any are coded, the intervening commas must be included so that the parser can determine which field is included.


qualifier1qualifier2

See notes on Qualifiers in “DTLDESCRIBE Qualifiers” on page 129. Returns the first and second data map qualifiers.

tablename This is the primary table qualifier. Returns the table name. All tables are returned if left blank.

comments If set to Y this will return any comments data found in the database.

ORDER This field must be left blank (comma needed to maintain position of later fields).

escape character Specifies the character to be used as the escape character (default is ~).

respect case 1respect case 2respect case 3

Set to Y this will respect the case of qualifier1, qualifier2 and tablename respectively - alternative is blank.

extended information If set to Y, returns the following fields containing extra information. Applicable to NRDB, NRDB2, and CAPX only. Defaults is N (no extended information required).

access method For NRDB,NRDB2, CAPX, and CAPXRT only:Specifies the valid access methods for which information is returned. The results of the DTLDESCRIBE are filtered to return information for only the specified access methods. This is useful to return a list of data maps that are applicable for certain access methods. Any number or combination of access method identifiers can be used (such as, OS or AKTM). If no value is used, the DTLDESCRIBE defaults to return all access methods.ADABAS - ACAPX / RT - CDATACOM - XDB2 - ZDB2390IMG - GDB2UNLD - WDL1 - DESDS - EIDMS - IISAM - MKSDS - KMQ - QODBA - ORRDS - NSEQ - STAPE - TUSER - UThe following extraction map access methods are applicable only to CAPX and CAPXRT:XDB2 - BXMSSQL - LXORACLE - P

Metadata Syntax 117

DTLDESCRIBE TABLES Returned Information

Column Name Description Extended information

qualifier_1 Creator/Schema name. See notes on Qualifiers in “DTLDESCRIBE Qualifiers” on page 129 for explanation.

No

qualifier_2 PowerExchange Data Map name will be returned, or null for NRDB2.

No

tablename Return table name. No

comments Comments column always present. Comment information is included only if requested by specifying the comments indicator in the describe.

No

type Type of object returned. I.e. TABLE No

Acc_mth The access method (from data map) Yes

Acc_mths01 ADABAS Database ID

SEQ, KSDS, ESDS MapFileName

IDMS MapSubSchemaName

ODBA or DL/I DBDName

Datacom Database ID

CAPX or CAPXRT The access method of the original data map

Other access methods NULL

Acc_mths02 ADABAS File Number

ODBA or DL/I DBDType

Datacom Table record length

IDMS or VSAM C (if table is valid for data capture) or NULL

CAPX Condense option (Full, Part or None)

DB2UNLD Null indicator value

Other access methods NULL

Acc_mths03 Datacom Table ID

IDMS or VSAM Database name

CAPX or CAPXRT Creator of the capture registration

DB2 DB2 Sub system ID

DB2UNLD Unload type

Other access methods Not Used. (comma needed to maintain position of later fields)

Acc_mths04 Datacom Version

CAPX or CAPXRT Name of the registration's source table

DB2 or DB2UNLD DB2 table name



DTLDESCRIBE PROCEDURESEnables you to list the available stored procedures:

dtldescribe procedures, [qualifier1], [qualifier2], [procedurename], [escape character], [respect case 1], [respect case 2], [respect case procname]

After the basic syntax of DTLDESCRIBE procedures all the other parameters [] are optional. If any are coded the intervening commas must be included so that the parser can determine which field is included.

DTLDESCRIBE PROCEDURES Returned Information

DTLDESCRIBE PROCEDURES is implemented for DB2400C and MSSQL.

DTLDESCRIBE PROCEDURECOLUMNSEnables you to list the available columns and information for given stored procedures:

dtldescribe procedurecolumns, [qualifier1], [qualifier2], [procedurename], [columnname], [escape character], [respect case 1], [respect case 2], [respect case procname], [respect case colname]

Acc_mths05 Datacom Recovery (Y/N)

DB2UNLD Unload file name


Base Record Describes the dependencies for a table. For example,ROOT(SEG2(SEG3,SEG4),SEG5)For Datacom, Base Record returns the short table name.


qualifier1 Catalog name.

qualifier2 Schema name.

procedurename Procedure name

escape character Specifies the character to be used as the escape character. Default is ~.

respect case 1respect case 2

Set to Y this will respect the case of qualifier1 / qualifier2 - alternative is blank.

respect case procname Set to Y this will respect the case of the procedure name - alternative is blank.

Column Name Description

PROCEDURE_CAT Procedure Catalog, null if not present; VARCHAR.

PROCEDURE_SCHEM Procedure Schema, null if not present; VARCHAR.

PROCEDURE_NAME Procedure Name, not null; VARCHAR.

NUM_INPUT_PARAMS Number of input parameters, not null (N/A for SQL Server); INTEGER.

NUM_OUTPUT_PARAMS Number of output parameters, not null (N/A for SQL Server); INTEGER.

NUM_RESULT_SETS Number of result sets, not null (N/A for SQL Server); INTEGER.

REMARKS Comments; VARCHAR.

PROCEDURE_TYPE Procedure type.

Column Name Description Extended information

Metadata Syntax 119

After the basic syntax of DTLDESCRIBE procedurecolumns all the other parameters [] are optional. If any are coded the intervening commas must be included so that the parser can determine which field is included.

DTLDESCRIBE PROCEDURECOLUMNS Returned Information


qualifier1 Catalog name.

qualifier2 Schema name.

procedurename Procedure name.

colname Column name.


respect case 1respect case 2

Set to Y to respect the case of qualifier1 / qualifier2 - alternative is blank.

respect case procname Set to Y this will respect the case of procedure name - alternative is blank.

respect case colname Set to Y this will respect the case of column name - alternative is blank.


PROCEDURE_CAT Procedure catalog, null if not present; VARCHAR

PROCEDURE_SCHEM Procedure schema, null if not present; VARCHAR

PROCEDURE_NAME Procedure name; VARCHAR

COLUMN_NAME Column name; VARCHAR

COLUMN_TYPE Column type, not null; SMALLINTCan be 0 - SQL_PARAM_TYPE_UNKNOWN1 - SQL_PARAM_INPUT - input parameter2 - SQL_PARAM_INPUT_OUTPUT - input/output parameter3 - SQL_RESULT_COLUMN - Parm is a column inset4 - SQL_PARAM_OUTPUT - output parameter5 - SQL_RETURN_VALUE - Column is return value of procedure

DATA_TYPE SQL datatype, not null; SMALLINT

TYPE_NAME Type name, character string representing datatype, not null; VARCHAR

COLUMN_SIZE Column size; INTEGER

BUFFER_LENGTH Buffer length. Maximum number of bytes required to store column data; INTEGER

DECIMAL_DIGITS Scale, NULL if not applicable; SMALLINT

NUM_PREC_RADIX Precision; SMALLINT

NULLABLE Nullable, not null, determines if column will accept a null value; SMALLINT. Can be0 - SQL_NO_NULLS1 - SQL_NULLABLE

REMARKS Remarks; VARCHAR

COLUMN_DEF Column default value; VARCHAR

SQL_DATA_TYPE SQL Datatype, not null; SMALLINT

SQL_DATETIME_SUB Datetime subtype. Can be1 - SQL_CODE_DATE2 - SQL_CODE_TIME3 - SQL_CODE_TIMESTAMP

CHAR_OCTET_LENGTH Maximum length in bytes of a character type column; INTEGER


DTLDESCRIBE PROCEDURECOLUMNS is implemented for DB2400C and MSSQL.

DTLDESCRIBE COLUMNSEnables you to list the available columns:

dtldescribe columns, [qualifier1], [qualifier2], [tablename], [comments], [ORDER], [escape character], [respect case 1], [respect case 2], [respect case 3], [extended information], [access method]

After the basic syntax of dtldescribe columns all the other parameters [] are optional. If any are coded, the intervening commas must be included so that the parser can determine which field is included.

ORDINAL_POSITION Ordinal Position if parameter given by Column name in result set, not null; INTEGER

IS_NULLABLE Can the column contain nulls. “YES” if column can contain nulls, “NO” if not; VARCHAR


qualifier1 For more information, see “DTLDESCRIBE Qualifiers” on page 129. Creator/Schema name.

qualifier2 Always NULL if describe is done using NRDB2 otherwise PowerExchange Data Map name.

tablename This is the primary table qualifier. Returns the table name. Columns for all available tables are returned if left blank.

Comments If set to Y this will return any comments data found in the database.

ORDER Optional argument, Y will force order by column number or blank will leave column order undefined.


respect case n Set to Y this will respect the case of qualifier1, qualifier2 and tablename respectively - alternative is blank.

extended information If set to Y returns additional information depending upon the data access type. Applicable to NRDB, NRDB2 and DB2 for z/OS and i5/OS only. Defaults to N (no extended information required).

access method For NRDB, NRDB2, CAPX, and CAPXRT only. Specifies the valid access methods for which information is returned. The results of the DTLDESCRIBE are filtered to return information for only the specified access methods. Use this method to return a list of data maps applicable for certain access methods. You can use any number or combination of access method identifiers. If you specify no value, the DTLDESCRIBE returns all access methods.ADABAS - ACAPX / RT - CDATACOM - XDB2 - ZDB2390IMG - GDB2UNLD - WDL1 - DESDS - EIDMS - IISAM - MKSDS - KMQ - QODBA - ORRDS - NSEQ - STAPE - TUSER - U


Metadata Syntax 121

DTLDESCRIBE COLUMNS Returned Information

Column Name DescriptionExtended Information

Qualifier1 Creator/Schema name. For more information, see Qualifiers in “DTLDESCRIBE Qualifiers” on page 129.

No

Qualifier2 PowerExchange data map name will be returned, or null for NRDB2.

No

Tablename Return table name. No

Column_name Column name. No

Type Field format. No

Precision Field length. No

Scale Decimal places. No

Nullable Nullable. No

Detail_type Internal column type representation. No

Comments Comments column always present. Comment information is included only if requested by specifying the comments indicator in the describe.

No

Key_type Reserved for future use. No

Ref_qualifier_1 Reserved for future use. No

Ref_qualifier_2 Reserved for future use No

Ref_table_name Reserved for future use No

Ref_column_name Reserved for future use No

Base Record Describes the dependencies for a table. For example, ROOT(SEG2(SEG3,SEG4),SEG5).

NRDB

Base Field Column base field, such as ROOT:ROOTKEY. NRDB

Base Field Type Field type of the column's base field (CHAR, VARCHAR etc.). NRDB

Base Field Offset Returns the offsets of the field within the record. There is a case when no offset is returned. If you have an array of fields and you collapse the array when you generate the table the offset of those fields will be 0.

NRDB

Field Usage Usage of field. For example, COMP or DISPLAY for Cobol. NRDB

Field Level Indent level of the field in the Cobol copybook. NRDB

Original Field Name Original name of the field as given in the Cobol copybook. NRDB

Field Picture Cobol picture format of the field. NRDB

CCSID CCSID. DB2

Internal_cp Internal code page. DB2, NRDB, NRDB2, CAPX, and CAPXRT

Codepage_name Code page name. DB2, NRDB, NRDB2, CAPX, and CAPXRT

Powercenter_name PowerCenter name of the code page. NRDB, NRDB2, CAPX, and CAPXRT


DTLDESCRIBE RECORDSEnables you to list the available columns. Applicable to NRDB and NRDB2 only.

dtldescribe records,[qualifier1], [qualifier2], [tablename], [comments], [ORDER], [escape character], [respect case 1], [respect case 2], [respect case 3], [extended information], [access method]

After the basic syntax of dtldescribe records all the other parameters are optional. If any are coded the intervening commas must be included so that the parser can determine which field is included.

DTLDESCRIBE RECORDS Returned Information

DTLDESCRIBE RECORDS returns fields containing the following information. All fields defaults to N which means that no extended information required.



For more information, see Qualifiers in “DTLDESCRIBE Qualifiers” on page 129.

tablename This is the primary table qualifier. Records for all available tables are returned if left blank.





extended information If set to Y returns additional information depending upon the data access type. Applicable to NRDB, NRDB2 and DB2 for MVS and AS400 only. Defaults to N (no extended information required).

access method For NRDB,NRDB2, CAPX, and CAPXRT only:Specifies the valid access methods for which information is returned. The results of the DTLDESCRIBE are filtered to return information for only the specified access methods. This is useful to return a list of data maps that are applicable for certain access methods. Any number or combination of access method identifiers can be used, such as OS or AKTM. If no value is used, the DTLDESCRIBE defaults to return all access method.ADABAS - ACAPX / RT - CDATACOM - XDB2 - ZDB2390IMG - GDB2UNLD - WDL1 - DESDS - EIDMS - IISAM - MKSDS - KMQ - QODBA - ORRDS - NSEQ - STAPE - TUSER - U

Field Returned information

Record name Name of the record.

Acc_mths01 IDMS - IDMS Record Name.

ODBA or DL/I - RecSegName.

Metadata Syntax 123

DTLDESCRIBE PKEYSdtldescribe pkeys, [qualifier1], [qualifier2], [tablename], [comments], [ORDER], [escape character], [respect case 1], [respect case 2], [respect case 3],[extended information],[access method]

After the basic syntax of dtldescribe pkeys all the other parameters are optional. If any are coded the intervening commas must be included so that the parser can determine which field is included.

Other access methods - NULL.

Acc_mths02 ADABAS - File Number.

ODBA or DL/I - DBDType.

Other access methods - NULL.

Acc_mths03 Not used. Reserved for future use. The comma is needed to maintain position of subsequent fields.

Acc_mths04 Not used. Reserved for future use. The comma is needed to maintain position of subsequent fields.

Acc_mths05 IDMS Log-Based. The technique that IDMS uses to physically store occurrences of the record type.C = CALCD = DIRECTV = VIAVS = VSAMVSC - VSAM CALC

Acc_mths06 IDMS - Record ID.

Acc_mths07 IDMS - Compressed.

Acc_mths08 IDMS - Variable.

Acc_mths09 IDMS - Page Group.

Acc_mths10 IDMS - Radix.

Acc_mths11 IDMS - Area Name.

Acc_mths12 IDMS - Minimum Root length.

Acc_mths13 IDMS - Data length.

Acc_mths14 IDMS - Prefix length.

Acc_mths15 IDMS - Control Portion length.

Source Filename Contains the file name of the Copybook. Not currently implemented.

DB Filename Contains the file name of the DBD source file. Not currently implemented.



For more information, see “DTLDESCRIBE Qualifiers” on page 129. Returns the first and second data map qualifiers.

tablename This is the primary table qualifier. Returns the table name.





Field Returned information


PKEYS is implemented for:

♦ ADABAS

♦ DB2

♦ DB2390IMG

♦ DB2400C

♦ DB2UDB

♦ MSSQL

♦ NRDB / NRDB2 (only for DL/I, ODBA, KSDS and Datacom data maps)

♦ ORACLE

DTLDESCRIBE FKEYSdtldescribe fkeys, [pk_qualifier1], [pk_qualifier2], [pk_tablename], [fk_qualifier1], [fk_qualifier2], [fk_tablename], [comments], [ORDER], [escape character], [respect case pk_1], [respect case pk_2], [respect case pk_3], [respect case fk_1], [respect case fk_2], [respect case fk_3],[extended information],[access method]

extended information This field must be left blank (comma needed to maintain position of later fields).

access method For NRDB, NRDB2, CAPX, and CAPXRT only:Specifies the valid access methods for which information is returned. The results of the DTLDESCRIBE are filtered to return information for only the specified access methods. This is useful to return a list of data maps that are applicable for certain access methods. Any number or combination of access method identifiers can be used, such as OS or AKTM. If no value is used, the DTLDESCRIBE defaults to return all access method.ADABAS - ACAPX / RT - CDATACOM - XDB2 - ZDB2390IMG - GDB2UNLD - WDL1 - DESDS - EIDMS - IISAM - MKSDS - KMQ - QODBA - ORRDS - NSEQ - STAPE - TUSER - U


pk_qualifier1pk_qualifier2

These refer to primary key qualifiers. For more information, see “DTLDESCRIBE Qualifiers” on page 129.

pk_tablename This is the primary table qualifier.

fk_qualifier1fk_qualifier2

These refer to foreign key qualifiers. For more information, see “DTLDESCRIBE Qualifiers” on page 129.

fk_tablename This is the foreign table qualifier.

comments Set to Y this will retrieve any comments data found in the database.

ORDER Must be blank (comma needed to maintain position if later arguments are used.).



Metadata Syntax 125

FKEYS is implemented for:

♦ DB2

♦ DB2UDB

♦ DB2400C

♦ ORACLE

♦ DB2390IMG

♦ MSSQL

General Notes

For metadata requests, the qualifiers will, as a default, be converted to the case that is standard on the source database, for example, uppercase for Oracle. Hence, supplying a qualifier1 field of scott or SCOTT will produce the same results because both will be treated as uppercase.

Alternatively, by setting the Respect case 1 option to Y, this defaulting to the database case will not be done. Hence a qualifier1 of scott will produce no results, whereas SCOTT will be successful.

Respect case 2 is relevant for qualifier2 and Respect case 3 is relevant for tablename.

The metadata qualifiers support wildcards as follows:

♦ * - One or more matching characters

♦ ? - A single matching character

If either of the wildcard characters are used within a column or table name, precede each occurrence of them with the escape character.

For example, a request for tab* would list all tables beginning with tab, whereas a request for tab~* would list only the table that was named tab*.

Therefore, to list only tables called s*ott in lowercase and return comments this SQL could be used:

dtldescribe tables,s~*ott,,,Y,,,Y

Table Information Retrieval

For example, to extract all the tables accessed by SCOTT, the following SQL would need to be supplied:

DTLDESCRIBE tables,SCOTT

Column Information Retrieval

For example, to extract all the column information about the EMP table owned by SCOTT, the following SQL would need to be supplied: Notice the use of the comma to maintain field position after the reserved field.

DTLDESCRIBE columns,SCOTT,,EMP

Primary Keys Information Retrieval

Table MFERNANDEZ.T4 has a primary key (PK4) defined on columns T41,T42.

DTLDESCRIBE pkeys,mfernandez,,t4

respect case pk_nrespect case fk_n

Set to Y this will respect the case of pk_qualifier1, pk_qualifier2 or fk_qualifier1, fk_qualifier2 and pk_tablename or fk_tablename as appropriate - alternative is blank.

extended information This field must be left blank (comma needed to maintain position of later fields).

access method This field must be left blank (comma needed to maintain position of later fields).



Result

MFERNANDEZ||T4|T41|1|PK4|MFERNANDEZ||T4|T42|2|PK4|

The numbers before PK4 are the order of the columns in the key definition.

Foreign Keys Information Retrieval

Example 1

Given a table, find all the foreign keys defined on that table.

Table MFERNANDEZ.T2 has two foreign keys.

(FK21) to T1 from column T21 to T11 with UPDATE rule RESTRICT and DELETE rule = DELETE

and

(FK23) to T3 from columns T22,T23 to T31,T32 with UPDATE rule NOACTION and DELETE rule = NOACTION

DTLDESCRIBE fkeys,,,,mfernandez,,t2

Result:

MFERNANDEZ||T1|T11|MFERNANDEZ||T2|T21|1|RESTRICT|RESTRICT|FK21|SQL010827171053710|MFERNANDEZ||T3|T31|MFERNANDEZ||T2|T22|1|NO ACTION|NO ACTION|FK23|PK3|MFERNANDEZ||T3|T32|MFERNANDEZ||T2|T23|2|NO ACTION|NO ACTION|FK23|PK3|

Example 2

Given a "parent table", find all the foreign keys that point to that table

DTLDESCRIBE fkeys,mfernandez,,t1

Result:

MFERNANDEZ||T1|T11|MFERNANDEZ||T2|T21|1|RESTRICT|RESTRICT|FK21|SQL010827171053710|

It is valid to query on both the parent and the foreign tables as shown below.

DTLDESCRIBE fkeys,mfernandez,,t*,mfernandez,,q*

DTLDESCRIBE SCHEMASEnables you to list the available schemas:

dtldescribe schemas, [filter_pattern],,,,,,,,,, [access method]

Metadata Syntax 127

After the basic syntax of dtldescribe schemas, the filter_pattern and access_method parameters are optional. All the other parameters are ignored. If any are coded the intervening commas must be included so that the parser can determine which field is included.

If no value is used, the DTLDESCRIBE defaults to return all access methods.

For example,

Dtldescribe schemas, d* ,,,,,,,,,, R

will return:

dev_schema but not eval_schema but only if dev_schema is using the CAPXRT access method.

This function is implemented for ADABAS Unload, CAPX, CAPXRT, DB2, DB2400C, IMSUNLD, INFORMIX, MSSQL, NRDB, NRDB2, ORACLE, SYBASE and DB2390IMG.


filter_pattern Supply a pattern to filter the schemas that are returned to just those that start with the supplied character. Use an asterisk (*) as a wildcard to match one or more characters of any type. For example, Dtldescribe schemas, d* returns:dev_schema but not eval_schema

access method For NRDB,NRDB2:Specifies the valid access methods for which information is returned. The results of the DTLDESCRIBE are filtered to return information for only the specified access methods. This is useful to return a list of data maps that are applicable for certain access methods. Any number or combination of access method identifiers can be used, for example, OS or AKTM. If no value is used, the DTLDESCRIBE defaults to return all access method.ADABAS - ACAPX / RT - CDATACOM - XDB2 - ZDB2390IMG - GDB2UNLD - WDL1 - DESDS - EIDMS - IISAM - MKSDS - KMQ - QODBA - ORRDS - NSEQ - STAPE - TUSER - U


DTLDESCRIBE Qualifiers

DB2 for i5/OS

DB2 for Linux, UNIX, and Windows

DB2 for z/OS

Informix

NRDB Nonrelational data formats, such as IMS, IDMS, Adabas, Datacom, VSAM, TurboIMAGE, and sequential files.

NRDB2 Nonrelational two-tier data formats.

Field Name Explanation

Qualifier 1 Optional, Schema name

Qualifier 2 Optional, Table owner


Qualifier 1 Optional, Schema

Qualifier 2 Optional, Definer (Creator ID)


Qualifier 1 Optional, Creator

Qualifier 2 Optional, DBName


Qualifier 1 Optional, Owner

Qualifier 2 Not Used


Qualifier 1 Optional, gives the first token of the data map name

Qualifier 2 Optional, gives the second token of the data map name


Qualifier 1 Optional, gives the first token of the data map name

Qualifier 2 Not used

DTLDESCRIBE Qualifiers 129

Oracle

Sybase


Qualifier 1 Optional, Schema name

Qualifier 2 Not used


Qualifier 1 Optional, Owner

Qualifier 2 Not Used


C H A P T E R 1 0

National Language Support


♦ Overview, 131

♦ Default Code Pages, 132

♦ General Code Pages, 132

♦ Code Pages for Data Sources, 133

♦ User-Defined Code Pages, 136

♦ ICUCHECK, 140

♦ Code Pages Shipped with PowerExchange, 147

♦ Handling Conversion Errors and Special Situations, 155

Overview

PowerExchange supplies default code pages based on International Components for Unicode (ICU) open source software.

PowerExchange uses the following type of code page specification:

♦ Default Code Pages. During PowerExchange installation, default code pages are installed in the installation directory for each PowerExchange platform.

♦ General Code Pages. PowerExchange uses the CODEPAGE configuration parameter to define the general code pages for transmitting or receiving data or metadata between platforms.

♦ Code Pages for Data Sources. Where possible, PowerExchange derives the code page of data from the database metadata. PowerExchange can use data source-specific configuration parameters to define the code pages for transmitting or receiving data or metadata between data sources.

♦ User-Defined Code Pages. Where it is not possible to derive the required code pages, or where data is loaded in a different code page to that which the database metadata expects, you must explicitly define the code pages for the databases that you use.

When the PowerExchange Listener starts up, it reports the code pages being used.

131

Default Code Pages

During PowerExchange installation, default code pages are installed in the installation directory for each PowerExchange platform.

The following table lists the default code page for each platform:

If these code pages are sufficient for your PowerExchange installation, do not configure any additional code pages. Typically, the default code pages are sufficient in the following circumstances:

♦ All data is held in single-byte code pages.

♦ Few accented characters are used.

♦ Data on databases and legacy files are in the same code page.

If the default code pages are not sufficient for your PowerExchange installation, use the CODEPAGE parameter in the DBMOVER configuration file to specify general code pages.

General Code Pages

Use general code pages if the default code pages that PowerExchange supplies are not sufficient for your installation. PowerExchange uses general code pages to transmit data or metadata between platforms.

Typically, general codes pages are used in the following circumstances:

♦ Multibyte data is processed.

♦ Data is outside of the ISO-8859 range on Linux, UNIX, or Windows, or outside of the IBM-037 range on i5/OS or z/OS.

Enter general code pages in the CODEPAGE statement of the DBMOVER configuration file. You can define the following types of general code pages in the CODEPAGE parameter:

♦ Control code page

♦ Data control code page

♦ SQL code page

The following table describes the uses of each code page type that can be specified in the CODEPAGE parameter:

Platform Code Page

i5/OS IBM-037

UNIX and Linux ISO-8859

Windows ISO-8859

z/OS IBM-037

Code Page Uses

Control code page - Names of databases, tables columns, or files.- User IDs or passwords.- Error messages.

132 Chapter 10: National Language Support

If you specify the control code page and omit the data control and SQL code pages, PowerExchange uses the control code page for the data control and SQL code pages. For example, the following statements are equivalent:

CODEPAGE=(UTF8) CODEPAGE=(UTF8,UTF8,UTF8)

If you omit the CODEPAGE parameter on Linux, UNIX, or Windows, PowerExchange uses the following default values:

CODEPAGE=(ISO-8859,ISO-8859,ISO-8859)

If you omit the CODEPAGE parameter on i5/OS or z/OS, PowerExchange uses the following default values:

CODEPAGE=(IBM-037,IBM-037,IBM-037)

Note: The PowerExchange Navigator overrides the code page specified in the DBMOVER configuration file, so that it can handle all characters. The PowerExchange Navigator uses UTF8 for the Control, Data, and SQL code pages.

Code Pages for Data Sources

The following table describes how the code page of a column is derived for each data source:

Data control code page - Column and parameter data used on nonrelational database types where the code pages have not been specified, for example, at the field or data map level.

- Parameter data refers to data sent to the server and substituted where parameter markers (question marks) are present in Delete, Select, or Update SQL.

SQL code page - Code page of SQL used to access data.- Typically, the SQL code page is the same as the control code page because table

names can be processed by both.- The SQL code page must be sufficiently precise to handle any literals.

Data Source Derivation of the Code Page

DB2 on i5/OS PowerExchange determines the internal code page number from the CCSID of the column and code page aliases.Optionally, columns that have no CCSIDs can be mapped to CHAR columns with code pages using the optional the DB2_BIN_CODEPAGE and DB2_BIN_AS_CHAR configuration parameters. Otherwise, these columns are mapped to BIN columns and can only be processed in hexadecimal format.

DB2 on z/OS Optionally, specify the DB2CODEPAGE parameter in the DBMOVER configuration file for each DB2 subsystem that serves as a PowerExchange data source.- PowerExchange determines the internal code page number of single-byte columns

from the CCSID of the column and the SCCSID of the EBCDIC_CCSID.- PowerExchange determines the internal code page number of double-byte (graphic)

columns from the CCSID of the column, and the GCCSID of the EBCDIC_CCSID.- PowerExchange determines the internal code page number of mixed byte columns

from the CCSID of the column, and the MCCSID of the EBCDIC_CCSID.

Nonrelational Bulk Data Movement

PowerExchange determines the internal code page number in the following order:1. The code page of the field from which the column was derived. This code page and field are specified in the data map.2. The code page of the data map.3. The CODEPAGE parameter for the data control code page, on the server where the NRDB access method runs.

Code Page Uses

Code Pages for Data Sources 133

DB2 for i5/OS Code Page ParametersSome systems define the data in tables as having a CCSID of 65535 (X'FFFF'). By default, PowerExchange treats such data as binary, which prevents the data from being translated. Binary data can be difficult to load onto a target system.

Optionally, set the following parameters in the DBMOVER configuration file to specify code page values for DB2 for i5/OS:

♦ DB2_BIN_AS_CHAR

♦ DB2_BIN_CODEPAGE

DB2_BIN_AS_CHAR

Use the DB2_BIN_AS_CHAR parameter to specify whether binary data is treated as character data.

DB2_BIN_AS_CHAR=Y¦N

DB2_BIN_CODEPAGE

Use the DB2_BIN_CODEPAGE parameter to associate the required single- and double-byte CCSIDs.

DB2_BIN_CODEPAGE=(SBCS_CCSID,DBCS_CCSID)

The following table describes the DB2_BIN_CODEPAGE parameters:

Note: PowerExchange supports ICU code pages only on DB2 for i5/OS 5.1 and later. For DB2 for i5/OS 4.5, PowerExchange uses static code pages and cannot access any column for which the CCSID maps to an ICU code page.

Use the DB2 access method, not the DB2400C access method, for code page support on DB2 for i5/OS.

Nonrelational Change Data Capture

PowerExchange determines the internal code page number in the following order:1. The code page of the field from which the column was derived. This code page and field are specified in the data map.2. The code page of the data map.3. The CODEPAGE parameter for the data control code page, on the server where the NRDB access method runs.PowerExchange records the code page of a field or data map in the CCT file when you create a capture registration.

Oracle Bulk Data Movement

PowerExchange determines the internal code page number from the character set portion of the NLS_LANG environment variable.Optionally, you can use the ORACLECODEPAGE parameter in the DBMOVER configuration file.

Oracle Change Data Capture

The PowerExchange internal code page for the number of columns from which changes are captured is always UTF-8.

Microsoft SQL Server PowerExchange determines the internal code page number from the collation sequence of the database.


SBCS_CCSID Specifies the single-byte character set.

DBCS_CCSID Specifies the double-byte character set.

Data Source Derivation of the Code Page


DB2 for z/OS Code Page ParametersYou can use the DB2CODEPAGE parameter in the DBMOVER configuration file to specify code page parameters for a DB2 for z/OS subsystem.

Determine the appropriate CCSID values for your DB2 subsystem by locating the installation job which created the DSNHDECP module.

Syntax

The DB2CODEPAGE parameter has the following syntax:

DB2CODEPAGE=(ssid, EBCDIC_CCSID=(SCCSID, GCCSID, MCCSID), ASCII_CCSID=(ASCCSID, AGCCSID, AMCCSID), UNICODE_CCSID=(USCCSID, UGCCSID, UMCCSID), MIXED=Y¦N )

You must include the SSID and EBCDIC_CCSID parameter. The ASCII_CCSID and UNICODE_CCSID parameters are optional.

For more information about DB2 CCSIDs, see the IBM DB2 for z/OS Installation Guide.

For each CCSID parameter, specify the following types of entries in the order shown:

1. SCCSID for single-byte data.

2. GCCSID for graphic data.

3. MCCSID for mixed-byte data.

You can define a maximum of 24 DB2COPEPAGE entries for one PowerExchange Listener.

If PowerExchange connects to a DB2 subsystem without a corresponding DB2CODEPAGE configuration entry, the data is processed with the default code page of the PowerExchange Listener. The PowerExchange Listener reports the code page being used.

If data is stored in more than one code page on DB2, or on other file systems that the PowerExchange Listener accesses, then you must specify a DB2CODEPAGE parameter for each DB2 subsystem accessed by PowerExchange.

Example

The following example indicates how to, look up the EBCDIC_CCSID CCSID values for Japanese Extended Katakana in the DB2 installation job for the DSNHDECP module, and then specify these CCSID values in the DB2CODEPAGE parameter.

The installation job contains the following CCSID values:

000273 //SYSIN DD *000274 DSNHDECM CHARSET=ALPHANUM,000275 ASCCSID=1041, X000276 AMCCSID=942, X000277 AGCCSID=301, X000278 SCCSID=290, X000279 MCCSID=930, X000280 GCCSID=300, X000281 USCCSID=367, X000282 UMCCSID=1208, X000283 UGCCSID=1200, X000302 END000303 //*

Code Pages for Data Sources 135

Enter these CCSID values in the PowerExchange in the DB2CODEPAGE parameter of the DBMOVER configuration file in the following manner:

DB2CODEPAGE=(D71G, EBCDIC_CCSID=(290,300,930), ASCII_CCSID=(1041,301,942), UNICODE_CCSID=(367,1200,1208))

Oracle Code Page ParameterOracle passes character data to PowerExchange in the character set component of the NLS_LANG environment variable. It is essential that the NLS_LANG environment variable is defined.

For many Oracle character sets, PowerExchange determines the code page with which to process connection strings, SQL statements, column data, and parameter data.

If PowerExchange cannot determine the code page that Oracle is using for character data, then you must define the ORACLECODEPAGE parameter in the DBMOVER configuration file to specify the code page. This parameter defines the PowerExchange and PowerCenter code pages that are used for an Oracle database. Use the following format:

ORACLECODEPAGE=(tnsname,pwxcp,pccp)

For example:

ORACLECODEPAGE=(BU920DTL,UTF-8,UTF-8)

The PowerExchange and PowerCenter code pages must match the environment variable NLS_LANG. You can specify up to 20 ORACLECODEPAGE parameters for separate databases or TNSNAMES.

User-Defined Code Pages

PowerExchange can use the following types of user-defined code pages:

♦ ICU-compatible code pages. These code pages are binary CNV files that are created from source definitions in a UCM file when you run the ICU makeconv utility. ICU supports all types of code pages including multibyte code pages.

♦ PowerExchange static code pages. These code pages are defined as a 16-by-16 matrix in text files. PowerExchange static code pages are restricted to single-byte code pages. Data cannot be converted to ICU code pages.

ICU-Compatible Code PagesUse the following task flow to add custom ICU code pages into PowerExchange code page internal slots:

To add a user-defined ICU code page:

1. Create the Unicode mapping (UCM) file, which defines the mapping between Unicode characters and the code page characters.

2. Run the ICU makeconv utility to build the run time CNV files on the platforms where PowerExchange will use the code pages.

3. Use the PowerExchange DBMOVER configuration parameters, ICUCONVERTER, ICUALIAS, and ICUDATADIR to add the new code page into an internal slot.

4. Perform a database row test from the PowerExchange Navigator to view the data.


Step 1. Create the UCM File

The UCM file maps the relationship between Unicode characters and hexadecimal values in the code page.

Edit the UCM file in a text editor. The file must contain only 7-bit ASCII characters on Linux, UNIX, and Windows platforms, or the equivalent characters on EBCDIC platforms.

The following table shows valid values for the precision indicator:

Most characters are defined as round-trip mappings, and have Unicode or hexadecimal values.

Sometimes, however, it is necessary to use one-way mappings, which cause an overlap between characters. For example, if a character is in the source code page but not in the target code page, or if a character is invalid, you can define a reverse-fallback mapping for it. When PowerExchange reads the data, the reverse-fallback mapping causes the character to be changed to a Unicode character that is a near match.

For more information about UCM file format, see http://icu.sourceforge.net/userguide/conversion-data.html#ucmformat.

Step 2. Run the ICU makeconv Utility

The makeconv utility converts a source code page definition from a UCM file into a binary CNV file.

The makeconv utility is an open source program that is available from the ICU Web site. For more information, see the ICU makeconv utility documentation at the following URL:

http://icu.sourceforge.net/userguide/icudata.html

The makeconv utility embeds the ICU version number into the CNV file name.

If you created custom code pages with earlier versions of ICU, you must regenerate the CNV files using ICU version 3.2.1.

Use the following syntax to run the makeconv utility and build a binary CNV file for the code page xxxx:

makeconv -p ICUDATA xxxx.UCM

You can use the makeconv utility on the following platforms:

♦ i5/OS. The utility runs as a *PGM from a SBMJOB command.

♦ UNIX and Linux

♦ Windows

♦ z/OS. The utility runs within the HFS UNIX system

Step 3. Add a Custom ICU Code Page

PowerExchange defines code pages internally using Code Page Numbers (CPNs). You can use the ICUCHECK utility to list the defined CPNs. User-defined ICU code pages must be defined in slots 301 to 340.

Precision Indicator

Description

0 Normal, round-trip mapping. A character retains the same hexadecimal value after being copied from the code page to Unicode and then copied back to the code page again.

1 Fallback mapping. A character is copied from Unicode to the code page but not back to Unicode again. If a character was copied back to Unicode again, it would get a new Unicode value.

2 Invalid character.The character will be replaced by the substitution character.

3 Reverse-fallback mapping.A character is copied from the code page to Unicode but not back to the code page. If the character was copied back to the code page again, it would get a different hexadecimal value.

User-Defined Code Pages 137

Add a custom code page on each platform that is involved in PowerExchange change data or bulk data movements. For example, if you are moving data from a data source on z/OS to a target on UNIX, you must add a code page definition to the DBMOVER configuration file on the z/OS system and on the UNIX system.

You must also add the new code page to the dbmover.cfg file on Windows so that you can view the data from the PowerExchange Navigator when performing a row test.

To add a custom ICU code page to an internal slot:

1. Move the CNV file to the required directories on each platform. The following table lists where CNV files are located for each platform:

2. Run ICUCHECK to determine which code pages are defined, and to verify that the current configuration and dynamic link libraries are correct. For more information, see “ICUCHECK” on page 140.

3. Add the ICUDATADIR parameter to the DBMOVER configuration file. In this file, specify the location of the CNV file.

4. Add an ICUCONVERTER(301,ADD...) parameter to the DBMOVER configuration file. Typically, the first user-defined ICU code page uses CPN 301, the next one uses 302, and so on.

5. If the code page is used by DB2 to map a CCSID to a code page, add an ICUALIAS=(301,ADD...) parameter to the DBMOVER configuration file.

6. If the code page replaces an existing code page that is used by DB2 to map a CCSID to a code page, add an ICUALIAS=(xxx,DELETE...) parameter the DBMOVER configuration file. Consult the ICUCHECK reports to determine the CPN of the existing code page.

7. Run ICUCHECK again to verify that the syntax of the PowerExchange configuration parameters is correct, and that the CNV file can be loaded. ICUCHECK does not start if the parameters contain syntax errors.

Step 4. View the Data from the PowerExchange Navigator

If the database type is nonrelational, define the code page at either the data map level for all CHAR and VARCHAR columns, or at the field level.

If the database type is DB2, verify that the CPxxx alias points to the new code page in CPN 301, and delete existing aliases that are pointing at a different code page.

To view the data from the PowerExchange Navigator:

1. In the Row Test dialog box, select Columns in Fetch and Extensions.

2. Run a row test.

3. Verify that the internal CPN number is 301.

4. In the Row Test dialog box, select Data in Fetch.

5. Run a row test of the data. The data can be accessed through Personal Metadata or an NRDB data map.

Platform Location of CNV Files

i5/OS UNIX style directory. The makeconv utility is run from the QSH shell.Set ICUDATADIR to the directory name.

UNIX and Linux

CNV files are located in the PowerExchange installation directory.If PowerCenter workflows are run on the machine, you must copy the CNV files to the server/bin directory which also contains the RES files.

Windows CNV files are located in the PowerExchange installation directory.If PowerCenter workflows are run on the machine, you must copy the CNV files to the server/bin directory which also contains the RES files.

z/OS UNIX style directory. The makeconv utility is run from USS.Set ICUDATADIR to the directory name.


6. Verify that the data is as expected.

PowerExchange Static Code PagesYou can define up to 10 external code page tables using fixed names. Code pages are defined in sequential text files that PowerExchange reads at run time. The parameter is specified in the following format:

CODEPAGE=(,USRCPNN,)

The variable NN is a number from 00 to 09, such as USRCP07.

Add a PowerExchange static code pages in the following manner:

♦ Add a control, data, or SQL code page in the DBMOVER configuration file.

♦ Add field or data map-level code pages in data maps for nonrelational database sources from the PowerExchange Navigator.

The following table specifies the code page file name and location that you must use on each platform:

You must specify 256 hexadecimal characters in 16 rows, each with 16 characters. Every character-pair of hexadecimal digits must be separated by a blank. The following example shows the code page table of an external code page file:

Platform Location and File name

i5/OS USRCPnn member that is specified in the CODEPAGE parameter of the DBMOVER configuration file.

UNIX and Linux USRCPnn.cp file in the directory where PowerExchange is installed.

Windows USRCPnn.cp file in the directory where PowerExchange is installed.

z/OS Data set that is allocated by the USRCPnn DD in the PowerExchange Listener or Client JCL.

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

40 4F 7F 7B 5B 6C 50 7D 4D 5D 5C 4E 6B 60 4B 61

F0 F1 F2 F3 F4 F5 F6 F7 F8 F9 7A 5E 4C 7E 6E 6F

AC C1 C2 C3 C4 C5 C6 C7 C8 C9 D1 D2 D3 D4 D5 D6

D7 D8 D9 E2 E3 E4 E5 E6 E7 E8 E9 AE BE 9E EC 6D

8C 81 82 83 84 85 86 87 88 89 91 92 93 94 95 96

97 98 99 A2 A3 A4 A5 A6 A7 A8 A9 8E BB 9C CC 00

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

00 AA B0 B1 9F B2 6A B5 BD B4 9A 8A BA CA AF BC

90 8F EA FA E0 A0 B6 B3 9D DA 9B 8B B7 B8 B9 AB

64 65 62 66 63 67 5A 68 74 71 72 73 78 75 76 77

7C 69 ED EE EB EF 5F BF 80 FD FE FB FC AD 4A 59

44 45 42 46 43 47 D0 48 54 51 52 53 58 55 56 57

79 49 CD CE CB CF A1 E1 70 DD DE DB DC 8D C0 DF

User-Defined Code Pages 139

To determine how the external code page file is specified:

1. Select a character in your local character set.

2. For the selected character, locate the equivalent character in the code page ISO-8859 and note the corresponding hexadecimal value.

3. Use this hexadecimal value to calculate the correct offset into the code page file that you are building.

4. At this offset, store the hexadecimal representation of your character.

For example:

1. Select 1 in your local z/OS character set. The character 1 is x'F1 in hexadecimal format.

2. Locate 1 in the code page ISO-8859 and find its corresponding hexadecimal value. This value is 31.

3. Locate offset x'31' in the code page file that is being defined. This offset location is at line 4, character 2, where the offset is relative to 1.

4. At that location, enter F1.

Because no mapping to Unicode exists, you cannot use any static code pages that you defined to perform the following tasks:

♦ View data by performing a row test in the PowerExchange Navigator.

♦ Convert characters to a target ICU code page.

ICUCHECK

ICUCHECK lists all of the code pages and aliases that are defined for a PowerExchange installation. The program checks the contents of the PowerExchange code page control table against the ICU data library and lists any differences.

Use ICUCHECK to perform the following tasks:

♦ List the contents of the code page control table.

♦ Verify that the correct version of the ICU data library is used, and that the path is correct.

♦ Verify that every ICU code page in the PowerExchange code page control table can be loaded, either from the ICU data library, or from a CNV file in the ICU data directory.

♦ Find the spelling of ICU converter names that you can copy and paste them into other areas such as, ODBC definitions on UNIX.

♦ Verify that the aliases are available for DB2 CCSID processing.

The output file contains reports that provide information about the available code pages. ICU provides the following reports:

♦ Report 1. Power Exchange Control Table for Simple Code Pages

♦ Report 2. PowerExchange Control Table for ICU Code Pages

♦ Report 3. Comparing Name and Character Size Information with ICU

♦ Report 4. Comparing ICU Code Page Information with PowerExchange

♦ Report 5. PowerExchange Code Page Names and Aliases

♦ Report 6. PMlocale Code Pages

♦ Report 7. Control Counts

♦ Report 8. Errors


SyntaxUse the following program to display information on the screen:

icucheck.exe

On Linux, UNIX, or Windows, use the following syntax to run the ICUCHECK program and write the results to a text file:

icucheck.exe > icucheck.txt

On i5/OS, use the SBMJOB command to run the ICUCHECK program. The reports are written to a QPRINT file.

On z/OS, add statements for running the ICUCHECK program to the JCL for a job. The reports are written to a SYSPRINT file.

Report 1. Power Exchange Control Table for Simple Code PagesThis report displays a list of all simple code pages. Simple code pages are static single-byte code pages that provide a fast translation facility with the following limitations:

♦ You can translate data between static code pages, but you cannot translate between static and ICU code pages.

♦ You can use only a one-to-one mapping between characters, however you cannot use a one-to-many or many-to-one mapping between characters.

Static single-byte code pages are defined in the range of 1 to 40. User-defined static code pages are defined in the range of 31 to 40.

When translating between code pages, such as from CPN 1 “ISO-8859” to CPN 3 “IBM-037,” the operating system builds a translation table of 256 bytes that contains the appropriate target values. To translate a particular value such as x'31', the system finds the value x'31' in the translation table and uses the translated value that is stored for it, x'F1'.

On platforms that do not support ICU processing, static code pages provide the facility to translate between code pages. Each static code page must have an equivalent ICU code page that is used if the other code page is ICU. For example, if you move data from CPN 3 “IBM-37” to CPN 41 “UTF-8,” the system converts from CPN 183 “ibm-37_P100-1995” to CPN 41 “UTF-8.”

This report contains the following columns:

Column Description

CPN PowerExchange code page number.

Name Name of the code page. PowerExchange and PowerCenter exchange data using this code page name.

PMlocale Name of the character set.

Power Center name Code page description.

Space Hexadecimal value of the space character used to pad CHAR columns.

CPN2 PowerExchange code page number of the equivalent ICU code page.

ICU Converter name Converter name of the equivalent ICU code page.

ICUCHECK 141

Sample Report

The following report is a sample PowerExchange control table for simple code pages report:

Report 1 : Power Exchange control table for Simple code pages=============================================================

NUM_STATIC_CODEPAGES = 41

CPN Name PMlocale Power Center name Space CPN2 ICU Converter name --- ---- -------- ----------------- ----- ---- ------------------ 001 ISO-8859 Latin1 ISO 8859-1 Western European 20 057 ISO-8859-1 002 IBM-1047 IBM1047 IBM EBCDIC US English IBM1047 40 213 ibm-1047_P100-1995 003 IBM-037 IBM037 IBM EBCDIC US English IBM037 40 189 ibm-37_P100-1995 ....

Report 2. PowerExchange Control Table for ICU Code PagesThis report displays details of all code pages known to PowerExchange.

The following code page number ranges are reserved for ICU code pages:

♦ The CPN range of 41 to 269. This code page range is used for the ICU code pages that PowerExchange supports, and that are present in the ICU data library, icudt32.dll.

♦ The CPN range of 270 to 300.

♦ The CPN range of 301 to 340.

To create a custom ICU code page by performing the following tasks:

1. Define the mappings in an ICU UCM file.

2. Run the makeconv utility to generate a binary CNV file from the UCM file.

3. Enter the name of the directory where the binary CNV file is located in the ICUDATADIR parameter of the DBMOVER configuration file.

4. Use the ICUCONVERTER=(301, ADD...) parameter in the DBMOVER configuration file to add the converter to the PowerExchange code page control table.

The report contains the following fields:

Field Description


Converter name ICU converter name.

Min Minimum bytes per character.

Max Maximum bytes per character.

In Increment added to the buffer size in addition to the standard formula. This increment allows space for leading escape sequences on code pages.

Al Number of aliases for this code page.

L Length of the space character.

Space Hexadecimal value of the space character that is used to pad CHAR columns.

Flag 1 : States Set to S if state bytes are used. For example:- X'0E', X'0F' in EBCDIC mixed-length code pages- Code pages with sub-versions such as ISO2022 and ISCII

Flag 2 : Endianness unknown

Set to U if the code page takes a different meaning depending on the endian format of integers. For example, UTF-16 on Windows means UTF-16LE, but UTF-16 on z/OS means UTF-16BE.

Flag 3 : ASCII / EBDIC Set to A if characters 0 - 9 and A-Z are compatible with 7-bit ASCII.Set to E if characters 0 - 9 and A-Z are compatible with standard EBCDIC, such as IBM-37.


Sample ReportReport 2 : Power Exchange control table for ICU code pages==========================================================

NUM_STATIC_CODEPAGES = 41number of ICU code pages = 300------------------------------NUMBER_OF_CODEPAGES = 341

Min : Minimum bytes per characterMax : Maximum bytes per characterIn : Increment to column size on top of standard formula column_size = to_max / (float)fr_min + 0.5)Al : Number of aliases

L : Length of the space characterSpace : Hex of the space character used to pad CHAR columns

Flag 1 : S = Are States UsedFlag 2 : U = Endianness UnknownFlag 3 : A = ASCII, E = EBCDICFlag 4 : P = SQL is parsable without translationFlag 5 : L = Lower case not standard for ASCII/EBCDIC SQLFlag 6 : X = Supplemental characters beyond UCS_2 plain

CPN Converter name Min Max In Al L Space Flags PMlocale Power Center Name--- -------------- --- --- -- -- - ----- ----- -------- -----------------041 UTF-8 1 3 0 10 1 20 A P X UTF-8 UTF-8 encoding of Unicode 042 UTF-16 2 2 2 0 2 0020 U X UTF-16 UTF-16 encoding of Unicode 043 UTF-16BE 2 2 0 11 2 0020 X UTF-16BE UTF-16 encoding of Unicode (Big Endian) ....

Report 3. Comparing Name and Character Size Information with ICUThis report displays the following information for each code page defined in the PowerExchange code page control table:

♦ Type of converter. This value is obtained from ICU.

♦ Canonical name of the converter. This value is obtained from ICU and is printed if it differs from the PowerExchange code page name.

♦ Minimum and maximum bytes for each character. These values are obtained from ICU. If they differ from the values used by PowerExchange, the difference is flagged.

If the code page is not present in the ICU data library DLL and no custom ICU code page CNV file occurs in the ICU data directory, U_FILE_ACCESS_ERROR is issued in an error message.

Flag 4 : Parsable without translation

Set to P if the invariant SQL characters are consistent with the standard for the ASCII / EBCDIC type. This flag is used within the ODBC layer to determine whether SQL is understandable on that platform. For example, if PowerCenter passes SQL in IBM EBCDIC Japanese CP939 to a Windows program compiled in code page ISO-8859, the SQL must be translated into an ASCII code page before key words like “select,” “from,” “where” can be recognized.

Flag 5: Lower case non-standard

Set to L if lowercase a to z characters are not consistent with standard EBCDIC such as IBM-37. This problem affects some Japanese code pages. Sometimes this problem is handled by forcing SQL into uppercase before converting it.

Flag 6: Supplemental Characters

Set to X if the code page includes supplemental characters beyond UCS_2.

PMlocale Name of the code page as it is understood by PowerCenter.

PowerCenter Name PowerCenter name of the code page. Data can be passed between PowerExchange and PowerCenter in this code page.

Field Description

ICUCHECK 143

The report contains the following fields:

Sample ReportReport 3 : Comparing PWX name, character size information with ICU==================================================================

ICU version = 3.2

If different from control array...Min = I18N_min_char_size resultMax = I18N_max_char_size resultCanonical name = I18N_canon_conv_name result

CPN Converter name Convert. type Min Max Supplementals--- -------------- ------------- --- --- -------------041 UTF-8 UTF8 contains supplementals042 UTF-16 UTF16 contains supplementals043 UTF-16BE UTF16_BigEnd. contains supplementals....

Report 4. Comparing ICU Code Page Information with PowerExchange

This report displays the number of ICU converters.

ICUCHECK.EXE makes iterative passes through all the code pages in the ICU data library using a method similar to the ICUINFO.EXE program. For each code page, it finds the matching entry in the PowerExchange code page control table and lists its CPN and PowerCenter name.

If no matching entry is found in PowerExchange, the following message is issued:

*** ICU converter not used

If a matching entry is found, the report contains the following columns:

Field Description



Convert. type Type of the ICU converter.

Min Minimum bytes for each character according to the ICU API, if different from the PowerExchange value.

Max Maximum bytes for each character according to the ICU API, if different from the PowerExchange value.

Supplementals Supplemental information.

Column Description


Min Minimum bytes for each character according to the ICU API, if different from the PowerExchange value.

Max Maximum bytes for each character according to the ICU API, if different from the PowerExchange value.

Converter type Type of ICU converter.



Sample ReportReport 4 : Comparing ICU code page information with PWX=======================================================

Number of ICU converters=235

Converter name Min Max Converter type CPN-------------- --- --- -------------- ---UTF-8 1 3 UTF8 041 UTF-16 2 2 UTF16 042 UTF-16BE 2 2 UTF16_BigEndian 043 ....

Report 5. PowerExchange Code Page Names and AliasesThis report displays the PowerExchange number associated with a code page. The report is arranged in ascending order by alias name.

PowerExchange uses aliases in the following situations:

♦ For DB2, PowerExchange adds the prefix “cp” in front of CCSID numbers to form aliases that are used to find the code pages. For example, a DB2 CCSID of 300 is changed to an alias of “cp300” before PowerExchange looks up the aliases to find PowerExchange CPN 252.

♦ You can specify the PowerExchange CODEPAGE configuration parameter with aliases, such as:

CODEPAGE=(IBM-37, CP930, CP930).

When using aliases in these situations, the “cp” format of alias or the PowerCenter name is recommended.

EBCDIC code pages are less common than ASCII. EBCDIC code pages usually have the literal “EBCDIC” in the alias. The following example shows aliases of Denmark EBCDIC Denmark code pages:

191 EBCDIC-CP-DK191 ebcdic-dk 225 ebcdic-dk-277+euro

Do not define the same alias on multiple code pages. Only the code page with the lower CPN number is used.

You can use the ICUALIAS parameter to change the alias definitions.

The report contains the following columns:

Sample ReportReport 5 : Power Exchange Code page Names and Aliases=====================================================

Number of Aliases = 1142

CPN Alias name Alias type--- ---------- ----------058 7-bit ASCII PM CP long description 171 Adobe Latin1 Encoding PM CP long description 171 AdobeLatin1Encoding 170 AdobeStandardEncoding ....

Title Description


Alias name ICU converter name.

Alias type Type of code page alias.

ICUCHECK 145

Report 6. PMlocale Code PagesThis report displays the PMlocale code pages.

The report contains the following columns:

Sample reportReport 6 : PM locale Code Pages===============================

id name description PWX codepages -- ---- ----------- ------------- 1 US-ASCII 7-bit ASCII US-ASCII 4 Latin1 ISO 8859-1 Western European ISO-8859,ISO-8859-1 5 ISO-8859-2 ISO 8859-2 Eastern European ibm-912_P100-1995 6 ISO-8859-3 ISO 8859-3 Southeast European ibm-913_P100-2000....

Report 7. Control CountsThis report displays summary statistics on the number of code pages by type.

Sample ReportReport 7 : Control counts=========================

CPN 001-030: Simple code pages 16CPN 031-040: User-defined Simple code pages 10CPN 041-300: ICU code pages 235CPN 301-340: User-defined ICU code pages 0----------------------------- ---CPN 001-340: Total code pages 261

Simple code pages with PMlocales 16ICU code pages with PMlocales 212----------------------------- ---Total code pages with PMLocales 228

Most aliases for a code page 11Code page with the most aliases 43 (UTF-16BE)

Power Center PMLocales not mapped to PWX 32....

Report 8. ErrorsThis report displays all errors and warnings.

Sample ReportReport 8 : Errors=================

PWX PMLocales not in Power Center : 3--------------------------------------------------- : ---Total errors : 3--------------------------------------------------- : ---

Return code set = 71 because errors were met

Column Description

id ID number of the code page.

name Name of the code page.

description Description of the code page.

PWX codepages Equivalent PowerExchange code page.


Code Pages Shipped with PowerExchange

To support different languages, the following types of code pages are shipped with PowerExchange:

♦ PowerExchange static single-byte code pages

♦ ICU version 3.2.1 code pages

♦ Additional Japanese ICU code page

ICU is an open source development project of IBM. For more information about ICU, see ICU Web site at:

http://icu.sourceforge.net/userguide/icudata.html

PowerExchange Static Code PagesThe following table lists the single-byte static code pages that are shipped with PowerExchange:

Note: Code pages beginning with "IBM-" are relevant to z/OS and i5/OS platforms.

Code Pages Holding All Unicode CharactersThe following table lists ICU code pages that handle all characters supported by Unicode:

Code Page Typical Use

ISO-8859 UNIX and Windows

IBM-037 USA, Canada, Brazil

IBM-273 Germany, Austria

IBM-277 Denmark, Norway

IBM-278 Finland, Sweden

IBM-280 Italy

IBM-284 Spain, Latin America

IBM-285 United Kingdom

IBM-297 France

IBM-424 Modern Hebrew

IBM-500 International

IBM-870 Polish

IBM-1047 Latin 1/Open Systems

MS-1250 PC Regional options setting for Central Europe (Polish)

Name Typical use Character Length in Bytes

BOCU-1 Email 3+

CESU-8 Compression 3+

IMAP-mailbox-name Email 3+

SCSU Compression 3+

UTF-16 2

UTF16_OppositeEndian 2

UTF16_PlatformEndian 2

UTF-16BE Databases 2

Code Pages Shipped with PowerExchange 147

Code pages UTF-16 and UTF-32 include a Byte Order marker (BOM) sequence of characters preceding the data. These code pages might be useful in encoding an entire file, but are seldom used in databases where UTF-16BE and UTF-16LE are commonly used or where characters are defined in 16 bit integers.

Little-ended (LE) code pages are used on Linux and Windows platforms, for example, UTF-16LE.

Big-ended (BE) code pages are used on Unix, i5/OS and z/OS, for example, UTF-16BE.

PowerExchange does not support supplemental characters defined above the basic plain, that is <U00FFFF>, and which are formed by 2 surrogate NUM16 integers. If PowerExchange encounters a supplemental character, PowerExchange treats it as invalid and replaces it with the substitution character.

Code Pages Based on ASCII EncodingThe following table lists ICU code pages that use the 7-bit ASCII values for the characters X'00' to X'7F':

UTF-16LE Databases 2

UTF-32 4

UTF32_OppositeEndian 4

UTF32_PlatformEndian 4

UTF-32BE 4

UTF-32LE 4

UTF-7 Email 3+

UTF-8 Databases 3+

Name Typical Use Character Length ins

ibm-1089_P100-1995 Arabic 1

ibm-1256_P110-1997 Arabic 1

ibm-5352_P100-1998 Arabic 1

windows-1256-2000 Arabic 1

ibm-1257_P100-1995 Baltic 1

ibm-5353_P100-1998 Baltic 1

ibm-901_P100-1999 Baltic 1

ibm-914_P100-1995 Baltic 1

ibm-921_P100-1995 Baltic 1

ibm-9449_P100-2002 Baltic 1

ibm-1131_P100-1997 Belarusian 1

ibm-1250_P100-1995 Central & East Europe 1




gb18030 Chinese 3+

ibm-964_P110-1999 Chinese 3+

HZ Chinese (simplified) 3+

ibm-1381_P110-1999 Chinese (simplified) 1 to 2

Name Typical use Character Length in Bytes




ibm-5478_P100-1995 Chinese (simplified) 2

windows-936-2000 Chinese (simplified) 1 to 2

ibm-1373_P100-2002 Chinese (traditional) 1 to 2



windows-950-2000 Chinese (traditional) 1 to 2

ibm-1251_P100-1995 Cyrillic 1



ibm-902_P100-1999 Estonian 1



ibm-1098_P100-1995 Farsi 1

ibm-1253_P100-1995 Greek 1

ibm-4909_P100-1999 Greek 1

ibm-5349_P100-1998 Greek 1

ibm-813_P100-1995 Greek 1

ibm-1255_P100-1995 Hebrew 1

ibm-5351_P100-1998 Hebrew 1

ibm-916_P100-1995 Hebrew 1

ibm-9447_P100-2002 Hebrew 1

ibm-1006_P100-1995 India 1

ibm-33722_P120-1999 Japanese 3+

ibm-33722_P12A-1999 Japanese 3+

ibm-942_P12A-1999 Japanese 1 to 2

ibm-943_P130-1999 Japanese 1 to 2

ibm-943_P15A-2003 Japanese 1 to 2

ibm-954_P101-2000 Japanese 3+

ibm-1363_P110-1997 Korean 1 to 2

ibm-1363_P11B-1998 Korean 1 to 2

ibm-949_P110-1999 Korean 1 to 2

ibm-949_P11A-1999 Korean 1 to 2

ibm-970_P110-1995 Korean 1 to 2

ibm-971_P100-1995 Korean 2

windows-1361-2000 Korean 1 to 2

windows-949-2000 Korean 1 to 2

ibm-1133_P100-1997 Lao 1



The official 7-bit ASCII characters are defined in code pages US-ASCII and ibm-367_P100-1995.

ICU Code Pages Based on EBCDIC Encoding

The following table lists ICU code pages that meet either of the following conditions:

♦ They support the characters a-z, A-Z, and 0-9 in the following standard EBCDIC ranges: x'81 to x'a9', x'c1' to x'e9, and x'f0 to x'f9.

♦ They are double-byte characters that are used with standard EBCDIC code pages.

iso-8859_10-1998 Swedish 1

ibm-1162_P100-1999 Thai 1

ibm-874_P100-1995 Thai 1

windows-874-2000 Thai 1

ibm-1254_P100-1995 Turkish 1

ibm-5350_P100-1998 Turkish 1

ibm-920_P100-1995 Turkish 1

ibm-1124_P100-1996 Ukrainian 1

ibm-1125_P100-1997 Ukrainian 1

ibm-1051_P100-1995 US and international 1




ISO-8859-1 US and international 1

US-ASCII US and international 1

ibm-1129_P100-1997 Vietnamese 1



Name Typical Use Character Length in Bytes

ibm-16804_X110-1999 Arabic 1

ibm-16804_X110-1999,swaplfnl Arabic 1

ibm-420_X120-1999 Arabic 1

ibm-918_P100-1995 Arabic 1

ibm-1112_P100-1995 Baltic 1

ibm-500_P100-1995 Belgium, Switzerland 1






ibm-837_X100-1995 Chinese (simplified) 2




ibm-1159_P100-1999 Chinese (traditional) 1


ibm-835_X100-1995 Chinese (traditional) 2






ibm-1142_P100-1997 Danish 1

ibm-1142_P100-1997,swaplfnl Danish 1

ibm-277_P100-1995 Danish 1

ibm-1137_P100-1999 Devanagari 1



ibm-1097_P100-1995 Farsi 1

ibm-1147_P100-1997 French 1

ibm-1147_P100-1997,swaplfnl French 1

ibm-297_P100-1995 French 1

ibm-1141_P100-1997 German 1

ibm-273_P100-1995 German 1

ibm-4971_P100-1999 Greek 1

ibm-875_P100-1995 Greek 1

ibm-12712_P100-1998 Hebrew 1

ibm-12712_P100-1998,swaplfnl Hebrew 1

ibm-424_P100-1995 Hebrew 1

ibm-4899_P100-1998 Hebrew 1

ibm-803_P100-1999 Hebrew 1

ibm-1149_P100-1997 Icelandic 1

ibm-1149_P100-1997,swaplfnl Icelandic 1

ibm-871_P100-1995 Icelandic 1

ibm-1144_P100-1997 Italian 1

ibm-1144_P100-1997,swaplfnl Italian 1

ibm-280_P100-1995 Italian 1



ibm-16684_P110-2003 Japanese 2








infa-FujitsuJEF Japanese 1 to 2

infa-FujitsuJEF_kana Japanese 1 to 2

infa-HitachiKEIS Japanese 1 to 2

infa-HitachiKEIS_kana Japanese 1 to 2

infa-jipse Japanese 1 to 2

infa-jipse_kana Japanese 1 to 2

infa-jp_ebcdic Japanese 1 to 2

infa-jp_ebcdik Japanese 1 to 2

infa-melcom Japanese 1 to 2

infa-melcom_kana Japanese 1 to 2

infa-unisys Japanese 1 to 2

infa-unisys_kana Japanese 1 to 2

ibm-13121_P100-1995 Korean 1

ibm-1364_P110-1997 Korean 1 to 2

ibm-833_P100-1995 Korean 1

ibm-834_P100-1995 Korean 2

ibm-933_P110-1995 Korean 1 to 2

ibm-1132_P100-1998 Lao 1

ibm-870_P100-1995 Polish 1

ibm-1153_P100-1999 Romanian 1

ibm-1153_P100-1999,swaplfnl Romanian 1

ibm-1145_P100-1997 Spanish 1

ibm-1145_P100-1997,swaplfnl Spanish 1

ibm-284_P100-1995 Spanish 1

ibm-1143_P100-1997 Swedish 1

ibm-1143_P100-1997,swaplfnl Swedish 1

ibm-278_P100-1995 Swedish 1

ibm-1160_P100-1999 Thai 1

ibm-838_P100-1995 Thai 1

ibm-1026_P100-1995 Turkish 1

ibm-1155_P100-1999 Turkish 1

ibm-1146_P100-1997 UK English 1

ibm-1146_P100-1997,swaplfnl UK English 1

ibm-285_P100-1995 UK English 1

ebcdic-xml-us US and international 1


ibm-1047_P100-1995,swaplfnl US and international 1



These code pages are based on IBM definitions and are used on i5/OS and z/OS.

IBM frequently describes code pages using CCSID numbers. PowerExchange uses aliases that begin with CP to match the CCSIDs. For example, "CP930" matches CCSID 930 and the ICU converter ibm-930_P120-1999.

IBM frequently indicates revisions by adding 4096 to the CCSID number, for example:

CCSID 300 and CCSID 16684 are related (16684 = 300 + 4096 * 4)

Each row in the following table lists CCSIDs that you can use in the same DB2 table on i5/OS or z/OS:

Specialized ICU Code PagesThe following table lists ICU code pages that are used in specialized contexts:









Typical UseSingle-ByteSCCSIDColumns

Double-ByteGCCSIDColumns

Mixed-LengthMCCSIDColumns

Japan - Extended English 1027 300 939

Japan - Extended Katakana 290 300 930

Japan - Katakana - Kanji 8482 16684 1390

Japan - Latin - Kanji 5123 16684 1399

Korean 833 834 933

Korean 13121 4390 1364

Simplified Chinese 836 837 935

Simplified Chinese 13124 4933 1388

Traditional Chinese 28709 835 937

Traditional Chinese (IBM Big-5) 1114 947 950


ibm-1276_P100-1995 Adobe 1

ibm-1277_P100-1995 Adobe 1

macos-0_2-10.2 Apple - US & International 1

macos-2566-10.2 Apple - East Asia 1 to 2

macos-29-10.2 Apple - Central & East Europe 1

macos-35-10.2 Apple -Turkish 1

macos-6-10.2 Apple - Greek 1

macos-7_3-10.2 Apple - Cyrillic 1



ibm-437_P100-1995 DOS 1

ibm-850_P100-1995 DOS 1

ibm-851_P100-1995 DOS -Greek 1

ibm-852_P100-1995 DOS 1

ibm-855_P100-1995 DOS - Cyrillic 1

ibm-856_P100-1995 DOS -Hebrew 1

ibm-857_P100-1995 DOS - Turkish 1

ibm-858_P100-1997 DOS 1

ibm-860_P100-1995 DOS - Portuguese 1

ibm-861_P100-1995 DOS - Icelandic 1

ibm-862_P100-1995 DOS - Hebrew 1

ibm-863_P100-1995 DOS - Canadian French 1

ibm-864_X110-1999 DOS - Arabic 1

ibm-865_P100-1995 DOS - Nordic 1

ibm-866_P100-1995 DOS - Russian 1

ibm-867_P100-1998 DOS - Hebrew 1

ibm-868_P100-1995 DOS 1

ibm-869_P100-1995 DOS - Greek 1

ibm-878_P100-1996 DOS - Russian 1

ibm-897_P100-1995 DOS - Japanese 1

ISCII,version=0 ISCII - Devanagari 3+

ISCII,version=1 ISCII - Bengali 3+

ISCII,version=2 ISCII - Gurmukhi 3+

ISCII,version=3 ISCII - Gujurati 3+

ISCII,version=4 ISCII - Orayi 3+

ISCII,version=5 ISCII - Tamil 3+

ISCII,version=6 ISCII - Telugu 3+

ISCII,version=7 ISCII - Kanada 3+

ISCII,version=8 ISCII - Malayalam 3+

ISO_2022,locale=ja,version=0 ISO2022 - Japanese 3+





ISO_2022,locale=ko,version=0 ISO2022 - Korean 3+

ISO_2022,locale=ko,version=1 ISO2022 - Korean 3+

ISO_2022,locale=zh,version=0 ISO2022 - Chinese 3+

ISO_2022,locale=zh,version=1 ISO2022 - Chinese 3+

LMBCS-1 Lotus Notes 1 to 2



Handling Conversion Errors and Special Situations

This section describes how to handle conversion errors and use a code page other than the one a database expects. This information is based on real-life situations.

Conversion ErrorsIf it is not possible to convert a character from one code page to another, ICU replaces the character with the substitution value for the current code page by default. Typically, this substitution value is a hexadecimal value, such as x'1A' or x'3F'.

The following situations can prevent character conversion:

♦ The character is not valid for the source code page. This situation can be caused by describing the data in the wrong code page.

♦ The character is valid in the source code page, but no equivalent character exists in the target code page. This situation is common if converting from multibyte characters to a single-byte target code page.

PowerExchange aborts when connecting to databases or describing tables if a code page conversion error occurs during initialization. PowerExchange uses substitution characters if any column or parameter data does not convert.

The following table summarizes conversion errors by context:

LMBCS-11 Lotus Notes - Thai 1 to 2

LMBCS-16 Lotus Notes - Japanese 1 to 2

LMBCS-17 Lotus Notes - Korean 1 to 2

LMBCS-18 Lotus Notes - Chinese 1 to 2

LMBCS-19 Lotus Notes - Chinese 1 to 2

LMBCS-2 Lotus Notes - Greek 1 to 2

LMBCS-3 Lotus Notes - Hebrew 1 to 2

LMBCS-4 Lotus Notes - Arabic 1 to 2

LMBCS-5 Lotus Notes - Cyrillic 1 to 2

LMBCS-6 Lotus Notes - Central & East Europe 1 to 2

LMBCS-8 Lotus Notes - Turkish 1 to 2

Data Type and Context Result Action

Names of databases, tables, columns, and files during initialization.

Processing aborts. Check the control code page.

User ID and password during connect processing.

Processing aborts. Check the control code page.

SQL during Describe processing.

Processing aborts. Check the SQL code page.


Handling Conversion Errors and Special Situations 155

Data in a Code Page That the Database Does Not ExpectIf you load data in a code page other than the one that the database metadata specifies, problems might occur when you try to convert data using PowerExchange or any conversion tool that the database provides. If the code page is already supported, you do not need to add a user-defined code page.

Use the ICUALIAS parameter in the DBMOVER configuration file on the system where the PowerExchange Listener runs, to replace the default internal code page number with the required internal code page number.

For example, if you need to consolidate data from different locations into DB2 for i5/OS tables. The CCSID is set to 37, but the data is in Chinese and Thai code pages. In this situation, extract the data to an Open Systems database target in UTF-8.

To extract the data to an Open Systems database target in UTF-8:

1. Create a separate PowerExchange Listener for each problem code page. Make sure that each PowerExchange Listener has a different i5/OS port and DBMOVER configuration file.

2. On the PowerExchange Listener that is used to access simplified Chinese data, set the ICUALIAS=(189,DELETE,CP37) configuration parameter to remove the default mapping between CCSID37 data and code page ibm-37_P100-1995.

3. On the PowerExchange Listener that is used to access simplified Chinese data, set the ICUALIAS=(208,ADD,CP37) configuration parameter so that DB2 describes the data using ICU code page ibm-935_P110-1999.

4. From the PowerExchange Navigator, run the DTLDescribe Columns command to verify that the data was described in the simplified Chinese code page.

5. From the PowerExchange Navigator, run a row test to view the data.

6. Use a PowerCenter workflow to copy the data to a flat file in UTF-8. Then view the flat file to verify that no characters were lost.

Column data during select processing.

Unrecognized characters are replaced with the substitution characters of the target code page which are hexadecimal values.Processing continues.

Check the data code pageVerify that the source data is described correctly using DTLDescribe columns.

Parameter data during delete, insert, select, and update processing.

Unrecognized characters are replaced with the substitution characters of the target code page which are hexadecimal values.Processing continues.

Check the data code page.Verify that the source and target data are described correctly using DTLDescribe columns.

Data Type and Context Result Action


C H A P T E R 1 1

Using the PowerExchange ODBC Driver


♦ PowerExchange ODBC Driver Data Source Parameters, 157

♦ Creating an ODBC Data Source on Windows, 164

♦ Creating an ODBC Data Source on UNIX, 169

♦ SQL Escape Sequences, 171

PowerExchange ODBC Driver Data Source Parameters

This section describes the PowerExchange ODBC driver parameters.

General PowerExchange ODBC Driver ParametersNote: PowerExchange ODBC driver data source parameters are case sensitive.

The following table lists the ODBC driver data source parameters:

Parameter Default Tokens Usage

CAPXIMAGETYPE NONE BA, AI or TU Enables the user to specify the capture image type when accessing PowerExchange Change Data Capture or PowerExchange Live data. - BA. Captures the Before Image (before the latest

change) and the After Image (the latest change). The Before Image is shipped with an action character forcing the deletion of the record.The After Image (which includes the change data) is in the format of an Insert.

- AI. Captures only the data After Image (the latest change).

- TU. Captures the data before and after images as they occur. The Before Image is shipped to enable the application to ensure nothing has changed in that record since the request. The After Image comes through as a change (or Update) record. The update is still a full image but the application picking this up will be able to change the target record rather than Delete it and Insert the new one as when using BA.

157

CAPXTYPE NONE SL or RS The xtype parameter can contain values depending on the extraction criteria. xtype defines the type of extract requested:- SL. Extract all data since last extract run.- RS. Restart previous (or specified) extract.

COMPATIBILITY NONE This should be left blank. It should only be used after contacting Informatica Global Customer Support.DFN returns the PowerExchange:data_source_type, for example:DETAIL:NRDB

Null returns just DETAIL.

COMPRESS N Y, N Compress data option.

CONFIRMWRITE Y Y, N - ConfirmWrite=Y. Ensures that data sent to the PowerExchange ODBC driver is immediately sent to the PowerExchange Listener (rather than buffered) and a success/no success response is received. This is important if good error recovery is a priority. It has the drawback of slowing data transfer rates. This is the default.

- ConfirmWrite=T. Enables asynchronous write with fault tolerance. For more information about synchronous write, see the PowerExchange Bulk Data Movement Guide.

- ConfirmWrite=N. Please contact Informatica Global Customer Support before using this setting. Applies to DB2, DB2 UDB, and Oracle only.

DATABASE For DB2 for Linux, UNIX and Windows. Only required if EMR is being used. This parameter overrides the database connection obtained from the extraction map.

DB2DEGREE None Any string of text up to 30 characters

Sends the SET CURRENT DEGREE SQL command to DB2, specifying whether or not, DB2 should use inter-partition parallel processing.Applies to DB2 for i5/OS and z/OS.The actual effect of each value can be found in the relevant IBM DB2 documentation detailing the “SET CURRENT DEGREE” and “CHGQRYA DEGREE” commands.For MVS only:Any text following the = is taken as text to follow a ‘SET CURRENT DEGREE =’. For example, specifying DB2DEGREE=ANY would execute the statement “SET CURRENT DEGREE = ‘ANY’” and would set DB2 to be able to use parallel processing.For i5/OS only:Any text following the = is taken as text to follow a ‘CHGQRYA DEGREE’ statement. For example, specifying DB2DEGREE=SYSTEM would execute the statement “CHGQRYA DEGREE(*SYSTEM)”. To maintain compatibility with previous releases, if the parameter begins with a digit, it is appended to a “CHGQRYA DEGREE(*NBRTASKS)” statement. Thus a “CHGQRYA DEGREE(*NBRTASKS 1)” statement can be generated with either a DB2DEGREE=1 or DB2DEGREE=”*NBRTASKS 1” statement.


158 Chapter 11: Using the PowerExchange ODBC Driver

DBQUAL1 None DBType

Filename ADAUNLD

i5/OS Library/Filename

CAPX and CAPX/RT

SSID or DBName

DB2, DB2400C, and DB2UDB

N/A DB2390IMG

Library/Filename EMR for i5/OS

N/A IMSUNLD

Database INFORMIX

DSN MSSQL

N/A NRDB and NRDB2

SQL *Net Name ORACLE

Server SYBASE

Sub-system Teradata

DBQUAL2 NONE DBType

Database ADAUNLD

Application name CAPX and CAPX/RT

N/A DB2, DB2400C, and DB2UDB

SSID DB2390IMG

N/A IMSUNLD

Server INFORMIX

Database MSSQL

N/A NRDB and NRDB2

N/A ORACLE

Database SYBASE

DBTYPE N/A Any supported data source, such as DB2

Access method for file or database.

DESCRIPTION NONE Text Descriptive text.

DRIVER N/A Path/File Path and file name of the PowerExchange ODBC driver. The file name can have the suffix.so or .sl according to the version of UNIX being run. Review the suffixes on the existing PowerExchange shared libraries to choose the correct one.

DTLCONN_OVR NONE N/A Specify the name in the CAP_CONNECTION statement in the DBMOVER configuration file to point to the relevant data source. For more information, see “Multiple CAPI Connections and Overrides” on page 42.


PowerExchange ODBC Driver Data Source Parameters 159

DTLIMGOV N Y or N For DB2390IMGIf the imagecopy data set is specified, PowerExchange checks that the imagecopy file exists and that a consistent imagecopy is selected. If either of these is not true (i.e. the imagecopy file does not exist or it is inconsistent) then the request is rejected and a message is displayed.This behavior can be overridden if required by supplying the parameter DTLIMGOV=Y. This enables inconsistent imagecopies to be read.

ENCRYPT N N, Y, DES or RC2

- N. No encryption is used when moving data.- Y. Data is encrypted by using a special Informatica

developed algorithm if it is being accessed by way of a PowerExchange Listener.

- DES. Data is encrypted by using Digital Encryption Standard (DES) algorithm. If this is selected it must be accompanied by a corresponding ENCRYPTLEVEL parameter.

- RC2. Data is encrypted by using the RSA security algorithm. If this is selected it must be accompanied by a corresponding ENCRYPTLEVEL parameter.

ENCRYPTLEVEL 1, 2 or 3 Must be used if the ENCRYPT parameter is set to DES or RC2.Valid values are:1. The data will be encrypted using:- For DES; 56 bit key encryption- For RC2; 56 bit key encryption2. The data will be encrypted using:- For DES; Triple DES 164 bit key encryption- For RC2; 64 bit key encryption3. The data will be encrypted using:- For DES; Triple DES 164 bit key encryption- For RC2; 128 bit key encryption

INTERPRETASROWS

N Y, N Specifies how PaceSize should be interpreted, either as rows or kilobytes.

ISOLATION ALL, CHG, CS NONE or RR

For DB2 for i5/OS, specifies the commit scope of the transaction. This is normally set to CS.For more information, see DB2 ISOLATION LEVEL.

JRNL Library/Journal Applies to i5/OS. The fully qualified library and journal name that is to be used instead of that specified in the PowerExchange Change Data Capture Registration.For example:STQA/NEWJOURNAL

LIBRARYLIST A maximum of 34 libraries may be specified each with a maximum length of 10 characters.

Space-delimited list of libraries that PowerExchange will search in order to qualify the first table name (if the table name is unqualified) on a select, insert, delete or update statement. Note: If both the library list override (LIBRARYLIST) and the database file override (OVRDBF) are specified, and a table exists in the LIBRARYLIST as well as in the OVRDBF, the OVRDBF takes precedence.



LOCALCODEPAGE None A number identifying the code page that is used. For example 41 represents UTF-8.If NONE is specified, at run time, the value from the DBMOVER configuration file will be used. A lookup list of numbers to code page name can be created using the icucheck.exe command-line utility. This utility is located in the root directory of your PowerExchange installation.It is recommended that the results of the utility be piped to a file for ease of lookup.For example:icucheck.exe > iculist.txt

LOCATION N/A NODE A NODE defined within the DBMOVER configuration file. For more information, see “DBMOVER Configuration File Parameters” on page 3.

LOGSID Enables you to specify the location of the PowerExchange log catalog. This is not available for use with the IDMS log-based change data capture environment on MVS.

MAXROWS Numeric Maximum limit for rows to be retrieved on this request. The maximum is 2,147,483,647 rows.Default is 2,147,483,647 rows.

MQGETOPT R B or R Indicates the type of read from a MQ Series queue:- B(rowse) reads a copy of the data on the queue.- R(ead) reads the data and removes it from the queue.

NOUPDATECDEP N Y/N If set to Y does not add the row test to the PowerExchange Change Data Capture audit trail for the application. This prevents cluttering up the audit trail with test data extractions.

ORACOLL Enables the PowerExchange Listener to submit multiple simultaneous Oracle Capture processes that can connect to different Oracle instances.Oracle Capture will use the override to determine to which instance it should connect, as opposed to using the ORACOLL keyword of the CAPI_CONNECTION TYPE=ORCL statement. in dbmover.cfg.This will allow the customer to use a single PowerExchange Listener to capture data from as many as 10 Oracle instances simultaneously.

ORACONN Enables the user to override Oracle connection information for a given Oracle Collection Id. This means that the user can use a single set of registrations to capture data from multiple Oracle instances.This overrides the fourth value in the ORACLEID statement in dbmover.cfg, for example:ORACLEID=(coll_id,oracle_sid, connect_string,cap_connect_string)Used in conjunction with the Oracle Instance.The user can specify either or both Instance/Connection string. If one of the keywords is not specified, Oracle Capture will pick up the value of the other from the dbmover.cfg file.



ORAINST Enables the user to override Oracle instance information for a given Oracle Collection Id. This means that the user can use a single set of registrations to capture data from multiple Oracle instances.This overrides the second value of the ORACLEID statement in dbmover.cfg, for example:ORACLEID=(coll_id,oracle_sid, connect_string,cap_connect_string)Used in conjunction with the Oracle Connection string. You can specify either or both Instance/Connection string. If one of the keywords is not specified, Oracle Capture will pick up the value of the other from the dbmover.cfg file

ORASCHEMA Enables the user to override the schema name to use for a group of registrations. This means that the user can use a single set of registrations to capture data from multiple schemas that may exist in a given Oracle instance

OVRDBF A maximum of 8 overrides can be specified. Each override can have a maximum length of 43 characters: 10 characters for each of library, file and member plus the three "/" separators.

Space-delimited list of i5/OS database file override. The format is: from_file/to_library/to_file/to_member

Where: - from_file. The file to be overridden.- to_library. The new library to use.- to_file. The file in the new library to use. - to_member. Optional. The member in the new library

and file to use. *FIRST is used if nothing is specified.

This override replaces any occurrence of the from_file name in any SQL statement with the to_library, to_file, and to_member. The override occurs whether the from_file is qualified or not. This includes any files that are qualified by using LIBRARYLIST.

Note: If both the library list override (LIBRARYLIST) and the file list override (OVRDBF) are specified, and a table exists in the LIBRARYLIST as well as in the OVRDBF, the OVRDBF takes precedence

PACESIZE 0 (zero) Numeric Number of rows or kilobytes. Resolves down to api rows_requested parm. Use when an application cannot keep up with the flow of data from the PowerExchange Listener (with interactive applications). For maximum performance the value should be zero.

REJECTFILE None Up to 384 characters

Specifies the location and file name of the Reject File that is used during Asynchronous Write.Only allowed if Write Mode is set to Asynchronous with Fault Tolerance. You can enter a value of PWXDISABLE to turn off reject file logging.For more information, see the PowerExchange Bulk Data Movement Guide.

STOPONERRORS 0 0 to 2147483647 Specifies the number of non-fatal errors that are allowed in the Write phase before processing is terminated.Only allowed if Write Mode is set to Asynchronous with Fault Tolerance. For more information, see the PowerExchange Bulk Data Movement Guide.



PowerExchange ODBC Driver Parameters for DB2 for z/OS Bulk Load Processing

The following PowerExchange ODBC driver parameters for DB2 for z/OS bulk load processing can be specified.

WAITTIME 0 & 2 - 86400 Sets the maximum approximate time (in seconds) to wait for data on a queue before returning EOF.- 0 indicates EOF should be returned as soon as the

end of the current log is reached. The end of the log is determined at the START of the extract process (as the real end is an ever moving target).

- 86400 indicates that EOF is never returned. The job will wait forever.

XTRASCHEMA Up to 128 characters with no spaces allowed.

This parameter overrides the schema defined in the extraction map.

Field Type Size Valid Values Comments

BULKLOAD Char 1 Y|N Enables or disables DB2 bulk load processing. Specify Y or N. If you specify Y, you can configure the following parameters.

CTLTEMPLATE Char 8 User Choice Name of the DB2 bulk load control file template on the target system. The DB2 bulk load control file template is a member in a PDS defined on the PowerExchange Listener platform.If CTLTEMPLATE is set to NONE, no DB2 bulk load control file template is specified. If CTLTEMPLATE is specified, LOADOPTIONS will be ignored.The default JCL is DB2LDJCL. You must customize this JCL before use.

DELETETEMPFILES Numeric int 1 Does not delete temporary files.

2 Deletes temporary files before running the loader.

3 Deletes temporary files after running the loader if the return code=0.Valid if MODETIME=1.

4 Deletes temporary files after running the loader.Valid if MODETIME=1.

FNAME Char 27 File on host Used as the stem for files created on MVS.The files created are:FNAME.DTLnnnnn - Data for LoadFNAME.DTLnnnnn.CTL - Load commandsFNAME.DTLnnnnn.SQL - SQL Template - Create TableFNAME.DTLnnnnn.SYSPRINT - Output from Loader



Creating an ODBC Data Source on Windows

Note: You cannot install the PowerExchange ODBC drivers on Microsoft Vista.

To create an ODBC data source:

1. In Control Panel, double-click Administrative Tools, and then double-click Data Sources (ODBC).

2. In the ODBC Data Source Administrator dialog box, click the System DSN tab.

3. Click Add.

4. In the Create New Data Source dialog box, select the Informatica PowerExchange driver from the list of available drivers and click Finish.

If this driver is not available, see the PowerExchange Installation Guide.

5. In the Name box, enter a name for the data source.

JCLTEMPLATE Char 8 User Choice Name of the JCL file template on the target system. This is a member in a PDS defined on the PowerExchange Listener platform. The default JCL is DB2LDJCL. You must customize this JCL before use.

LOADOPTIONS 1 or 4 INSERT (1) or REPLACE (4).

MODETIME Numeric int 1 = WAIT Synchronous - report results at end. Applicable only if MODETYPE=1 or MODETYPE=3 is specified.

2 = NOWAIT Asynchronous - returns OK when job is submitted. Applicable only if MODETYPE=3 or MODETYPE=2 is specified.

3 = TIMED Synchronous - only wait a specified number of seconds. Applicable only if MODETYPE=3 is specified.

4 = DATAONLY

Only creates the data DB2 Loader file. Applies to all Mode Type values but normally only used with NOSUBMIT, MODETYPE=2.

MODETYPE Numeric int 1 = TASK SubTask under the PowerExchange Listener

2 = NOSUBMIT

This option will create the DB2 Loader specific control files and data files, save them at the target and generate a executable JCL deck on disk.

3 = JOB Submitted to the internal reader as a job.

PRISPACE Numeric int 1-System defined

Amount of primary space for MVS files.

SECSPACE Numeric int 1-System defined

Amount of secondary space for MVS files.

SPACE Char 1 T = Tracks C = Cylinders

MVS allocation of space.

TIME Numeric int 0-99999 0 = NOWAIT, 99999=WAIT any other is number of seconds. Only used in conjunction with MODE=(JOB,TIMED).

Field Type Size Valid Values Comments


6. In the Location list, click the location. The location is defined in a NODE statement in the DBMOVER configuration file. For more information, see “DBMOVER Configuration File Parameters” on page 3.

7. In the Type list, click a data source type.

8. Depending on the data source type that you selected, enter values for all of the available properties.

9. In the Local Codepage list, click a codepage. If you click Default, at run time, the value from the PowerExchange configuration file or the default for the platform is used.

10. Click OK.

The ODBC data source appears in the System Data Sources list in the ODBC Data Source Administrator dialog box.

11. Click OK.

PowerExchange Data Source TabThis tab enables you to enter the minimum details required to create a data source. You can enter the following information:

Field Corresponding ODBC Parameter

Name Entry in ODBC.ini file

Location LOCATION

Type DBTYPE

Local Codepage This listbox contains all the available code pages. The default is "Default" which means that at runtime the value from dbmover.cfg (or the default for the platform) will be used.

The following fields are displayed according to the data source that is selected in Type:

ADAUNLD

File Name DBQUAL1

Database DBQUAL2

CAPX

AS400 Lib/File i5/OS library and file name

Application Name DBQUAL2

CAPXRT


Wait Time WAITTIME

DB2

SSID or DBName DBQUAL1

Write Mode CONFIRMWRITE

Reject File REJECTFILE

Stop On Errors STOPONERRORS

DB2390IMG

SSID DBQUAL1

DB2400C



DB2UDB

Creating an ODBC Data Source on Windows 165

General TabOn the General tab, set optional properties. You can enter the following information:





EMR


Wait Time WAITTIME

INFORMIX or SYBASE

Database DBQUAL1

Server DBQUAL2

MSSQL

DSN DBQUAL1

Database DBQUAL2


NRDB and NRDB2


Wait Time WAITTIME

ORACLE

SQL *Net Name DBQUAL1





Integration Mask COMPATIBILITY

Maximum Rows MAXROWS

Pacing Size PACESIZE

As KB / As Rows INTERPRETASROWS

Compress COMPRESS

Encrypt If unchecked then Encrypt = NIf checked and Option left blank then Encrypt = Y

Option If set to DES then Encrypt = DES. Level must then be set.If set to RC2 then Encrypt = RC2. Level must then be set.

Level ENCRYPTLEVEL

MQ Get Opt MQGETOPT

Application DTLAPPApplication name.The application name is used as part of the audit trail for data extractions by the PowerExchange Listener.



DB2/S390 Bulk Load Properties TabThis tab applies to DB2 data sources only. It enables you to enter bulk load processing properties. You can enter the following information:

AS/400 Properties TabThis tab is for DB2 for i5/OS only. You can enter the following information:

EMR Properties TabThis tab is available only if you selected EMR in the data source Type field. You can enter the following information:


Bulk Load BULKLOAD. If this is Y (i.e. checked) enables the subsequent parameters for input.

Load Options LOADOPTIONS

Delete Temporary Files DELETETEMPFILES

JCL Template JCLTEMPLATE

CTL Template CTLTEMPLATE

Mode Type MODETYPE

Mode Time MODETIME

Time TIME

Space SPACE

Primary Space PRISPACE

Secondary Space SECSPACE

FName FNAME


Isolation Level ISOLATION

Library List (Unqualified Table) LIBRARYLIST

Table/File Override OVRDBF


Extract CAPXTYPE

Image CAPXIMAGETYPE

AS400 Library/Filename DBQUAL1

AS400 Library/Journal JRNL

Oracle Instance ORAINST

Oracle Connection String ORACONN

Oracle Collection ID ORACOLL

HP3000 Volume Set No longer supported.

DB2 UDB Database DATABASE

Creating an ODBC Data Source on Windows 167

CAPX Properties TabThis tab is only available if you have selected CAPX in the Data Source Type field. You can enter the following information:

CAPXRT Properties TabThis tab is available only if you selected CAPXRT in the data source Type field. You can enter the following information:


Extract CAPXTYPE

Image CAPXIMAGETYPE


AS400 Instance DBQUAL2

CAPI_Connection DTLCONN_OVRUsed to override the default database connection made in the DBMOVER configuration file. Specify the value given in the NAME parameter of the required CAPI_CONNECTION statement in the DBMOVER configuration file to point to the relevant data source. For more information, see “Multiple CAPI Connections and Overrides” on page 42.

Extraction Schema XTRASCHEMAEnables you to specify a schema which will override the schema specified in the extraction map.


Extract CAPXTYPE

Image CAPXIMAGETYPE


AS400 Library/Journal JRNL

Oracle Instance ORAINST

Oracle Connection String ORACONN

Oracle Schema ORASCHEMA

DB2 UDB Database DATABASE

CAPI_Connection DTLCONN_OVRUsed to override the default database connection made in the DBMOVER configuration file. Specify the value given in the NAME parameter of the required CAPI_CONNECTION statement in the DBMOVER configuration file to point to the relevant data source. For more information, see “Multiple CAPI Connections and Overrides” on page 42.

Extraction Schema XTRASCHEMAEnables you to specify a schema which will override the schema specified in the extraction map.


Accessing Multibyte Metadata with the PowerExchange ODBC DriverUse the following procedure to access multibyte metadata with the PowerExchange ODBC driver.

To access multibyte metadata with the Informatica PowerExchange Unicode driver:

1. In Control Panel, double-click Administrative Tools, and then double-click Data Sources (ODBC).

2. In the ODBC Data Source Administrator dialog box, click the System DSN tab.

3. Click Add.

4. In the Create New Data Source dialog box, select the Informatica PowerExchange driver from the list of available drivers and click Finish.

If this driver is not available, see the PowerExchange Installation Guide.

5. On the PowerExchange Data Source tab, enter a name for the data source in the Name box.

6. In the Local Codepage list, select the Unicode for metadata driver.

7. Click OK.

8. Click OK.

Creating an ODBC Data Source on UNIX

If the ODBC layer is required on UNIX, update the appropriate odbc.ini file.

The PowerExchange ODBC driver parameters are described in “General PowerExchange ODBC Driver Parameters” on page 157 and “PowerExchange ODBC Driver Parameters for DB2 for z/OS Bulk Load Processing” on page 163.

Note: It is important to set the ODBCINI environment variable to specify where the file is located. This variable must specify the full path and file name.

Example PowerExchange Data Source SpecificationsThe PowerExchange ODBC driver data sources are defined in the ODBC Data Sources section and then the PowerExchange ODBC driver parameters are set in the individually named section.

For example the infadb2 data source parameters are defined in the [infadb2]section:

[ODBC Data Sources] infadb2=INFA PWX850 DB2infanrdb=INFA PWX850 NRDBinfasybase=INFA PWX850 SYBASEloaddb2=INFA PWX850 DB2LOADoracleunicode=INFA PWX850 UNICODE

[infadb2]DRIVER=/usr/powerexchange/bin/libdtlodbc.sl DESCRIPTION='Descriptive Text for DB2 Data Source section'DBTYPE=db2 LOCATION=db2 DBQUAL1=dsn1

[infanrdb]DRIVER=/usr/powerexchange/bin/libdtlodbc.soDESCRIPTION='Freeform text for nrdb section'DBTYPE=nrdb LOCATION=laptop

[infanrdb2]

Creating an ODBC Data Source on UNIX 169

DRIVER=/usr/powerexchange/bin/libdtlodbc.soDESCRIPTION='Freeform text for nrdb2 section'DBTYPE=nrdb2 LOCATION=node1

[infasybase]DRIVER=/usr/powerexchange/bin/libdtlodbc.slDESCRIPTION='This accesses myserv Sybase on my PC'DBTYPE=sybase LOCATION=mypcDBQUAL1=myserv

[loaddb2]DRIVER=/usr/lw/stprod310/libdtlodbc.soDESCRIPTION=LoadDB2DBTYPE=DB2LOCATION=RemoteLOC (From DBMOVER.CFG)DBQUAL1=SSID (target database SSID)BULKLOAD=YCTLTEMPLATE=DB2LDCTLJCLTEMPLATE=DB2LDJCLDELETETEMPFILES=2 (BEFORE)FNAME=INFA.V310.ODBCLOADOPTIONS=4 (REPLACE)MODETIME=1 (WAIT)MODETYPE=3 (JOB)PRISPACE=1SECSPACE=1SPACE=tTIME=0

[oracleunicode]DRIVER=/dtlqa2/v850/libdtlodbcu.soDESCRIPTION=DATABASE=DBTYPE=ORACLELOCATION=tortillaDBQUAL1=TO101UTFMAXROWS=0COMPRESS=NENCRYPT=NCONFIRMWRITE=NPACESIZE=0INTERPRETASROWS=NBULKLOAD=NDELETETEMPFILES=0LOADOPTIONS=0LOCALCODEPAGE=41MODETYPE=0MODETIME=0TIME=0

Note: If no ODBC Driver Manager is installed, you can use the one provided with the PowerExchange software. This code is provided as libdtlodbcinst.so or libdtlodbcinst.sl (HP only) so as not to impact any existing ODBC Driver Manager. In order to use this code, you must either:

♦ Rename it to libodbcinst.so or libodbcinst.sl (HP only).

♦ Create a hard or symbolic link to it as libodbcinst.so or libodbcinst.sl (HP only).

Accessing Multibyte Metadata with ODBCYou must use the Informatica PowerExchange Unicode driver and local code page 41 to access multibyte metadata. In the odbc.ini file example in “Example PowerExchange Data Source Specifications” on page 169, the oracleunicode data source is defined and the following parameters set:

DRIVER=/dtlqa2/v850/libdtlodbcu.soLOCALCODEPAGE=41


SQL Escape Sequences

SQL Escape sequences can be used to override PowerExchange defaults.

SQL Escape Sequences with PowerCenterIn the PowerCenter Workflow Manager, you can enter SQL escape sequences for a particular task.

Note: If you enter multiple escape sequences, enclose each one in curly brackets.

For example:

{DTLXTYPE=RS}{DTLIMTYPE=BA}{DTLTIMEOUT=60}d8robora1

For more information, see PowerExchange Interfaces for PowerCenter.

SQL Escape SequencesThe following SQL escape sequences are available:

♦ {DTLADAPREFETCH=Y|N}, 172

♦ {DTLADAPWD= }, 172

♦ {DTLAPP= }, 172

♦ {DTLCONNOVR= }, 172

♦ {DTLCONFWRITE=N|Y|T}, 172

♦ {DTLDATAPWD= }, 172

♦ {DTLDB2DEGREE= }, 172

♦ {DTLDSN= }, 173

♦ {DTLEVENTTB= }, 173

♦ {DTLIMTYPE=BA|AI|TU}, 173

♦ {DTLIMGOV=Y|N}, 173

♦ {DTLINSMODE=LOAD|UPDATE}, 174

♦ {DTLJRNL= }, 174

♦ {DTLLIBRARYLIST= }, 174

♦ {DTLNOUPDATECDEP=Y|N}, 174

♦ {DTLORACOLL= }, 174

♦ {DTLORACONN= }, 174

♦ {DTLORAINST= }, 174

♦ {DTLORASCHEMA= }, 175

♦ {DTLOVRDBF= }, 175

♦ {DTLREJECTFILE= }, 175

♦ {DTLSESSID=}, 175

♦ {DTLSTOPAFTER=}, 175

♦ {DTLSTRIPORDERBY=Y|N}, 175

♦ {DTLTIMEOUT= }, 176

♦ {DTLUDBDB= }, 176

♦ {DTLUPDELSEINS=Y}, 176

♦ {DTLWORKERS= }, 176

♦ {DTLXTRASCHEMA= }, 176

SQL Escape Sequences 171

♦ {DTLXTYPE=RS|SL}, 176

{DTLADAPREFETCH=Y|N}Adabas only. Enables users to set prefetch for ADABAS.

{DTLADAPWD= }Adabas only. Enables users to specify an ADABAS file password as an escape sequence in ODBC.

{DTLAPP= }Enables you to specify an extraction application name without defining a new data source when you are using ODBC to access PowerExchange Change Data Capture.

{DTLCONFWRITE=N|Y|T}Overrides the Write Mode.

♦ N = Confirm Write Off

♦ Y = Confirm Write On

♦ T = Asynchronous Write with Fault Tolerance.

For more information about asynchronous write, see the PowerExchange Bulk Data Movement Guide.

{DTLCONNOVR= }Overrides the default CAPI connection specified in the DBMOVER configuration file. Specify the value given in the NAME parameter of the required CAPI_CONNECTION statement in the DBMOVER configuration file to point to the relevant data source. For more information, see “Multiple CAPI Connections and Overrides” on page 42.

{DTLDATAPWD= }Where xxxx is a password needed to access the physical data file on a machine. Enables users to specify a file password as an escape sequence in ODBC. This operates the same way as DTLADAPWD and in fact is the preferred keyword with DTLADAPWD being kept as a synonym for backward compatibility. Used with NRDB and NRDB2 access methods.

{DTLDB2DEGREE= }DB2 for i5/OS and z/OS only. Sends the SET CURRENT DEGREE SQL command to DB2, specifying whether or not DB2 should use inter-partition parallel processing. Can be any text string up to 30 characters.

The actual effect of each value can be found in the relevant IBM DB2 documentation for the “SET CURRENT DEGREE” and “CHGQRYA DEGREE” commands.

DB2 for z/OS

Any text following the = is taken as text to follow a ‘SET CURRENT DEGREE =’.

For example, specifying DB2DEGREE=ANY would execute the statement “SET CURRENT DEGREE = ‘ANY’” and would set DB2 to be able to use parallel processing.

DB2 for i5/OS

Any text following the = is taken as text to follow a ‘CHGQRYA DEGREE’ statement.


For example, specifying DB2DEGREE=SYSTEM would execute the statement “CHGQRYA DEGREE(*SYSTEM)”.

To maintain compatibility with previous releases, if the parameter begins with a digit, it is appended to a “CHGQRYA DEGREE(*NBRTASKS)” statement. A “CHGQRYA DEGREE(*NBRTASKS 1)” statement can be generated with either a DB2DEGREE=1 or DB2DEGREE=“*NBRTASKS 1” statement.

{DTLDSN= }Enables SQL to override the physical file name coded in the data map. This enables users to use ODBC to read NRDB and NRDB2 data from various files through one ODBC data Source. For ODBC, this is coded anywhere in the SQL as {DTLDSN=xxx}.

TABLE="{DTLDSN=filename}NRDBSchema.NRDBMapName.NRDBTableName"TABLE="{DTLDSN=filename}DB2Schema.DB2Table"

♦ Filename is any valid filename for the platform on which it resides. For NRDB and NRDB2 the filename will override the file name specified in the PowerExchange data map NRDBSchema.NRDBMapName.NRDBTableName.

♦ For the DB2 bulk load, filename will override the value FName specified in the ODBC setup (ODBC.INI).

♦ For i5/OS, the fully-qualified library and table name to be used instead of that specified in the PowerExchange change data capture registration. For example:

{DTLDSN=STQA/NEWTABLE}

If tables exist in multiple libraries you can define this by using an asterisk (*) instead of the library name. Matching tables in all libraries will be used.

For example:

{DTLDSN=*/NEWTABLE}

{DTLEVENTTB= }Activates Event Processing. Supply the name of the extraction map of the Event Table.

{DTLIMTYPE=BA|AI|TU}Enables users to specify the capture image type when accessing PowerExchange Change Data Capture through ODBC:

♦ BA. Delivers the Before Image (before the latest change) and the After Image (the latest change). The Before Image is shipped with an action character forcing the deletion of the record. The After Image (which includes the change data) is in the format of an Insert.

♦ AI. Delivers only the After Image data for the update.

♦ TU. Delivers the data before and after images as they occur. The Before Image is shipped to enable the application to ensure nothing has changed in that record since the request. The After Image comes through as a change (or Update) record. The update is still a full image but the application picking this up will be able to change the target record rather than Delete it and Insert the new one as when using BA.

{DTLIMGOV=Y|N}DB2 image copy access method (DB2390IMG) only. If the imagecopy data set is specified, PowerExchange checks that the imagecopy file exists and that a consistent imagecopy is selected. If either of these is not true, for example, the imagecopy file does not exist or it is inconsistent, then the request is rejected and a message is displayed.

This behavior can be overridden if required by supplying the parameter {DTLIMGOV=Y}. This enables inconsistent imagecopies to be read.


{DTLINSMODE=LOAD|UPDATE}The purpose of this escape sequence is to enable empty VSAM files to be loaded and updated. LOAD is the default and requires that all the operations be INSERTs. If the input file includes both inserts and then updates to those inserted records specify DTLINSMODE=UPDATE.

{DTLJRNL= }i5/OS only. The fully qualified library and journal name that is to be used instead of that specified in the PowerExchange DBMOVER configuration file.

For example:

{DTLJRNL=STQA/NEWJOURNAL}

{DTLLIBRARYLIST= }i5/OS only. Space-delimited list of libraries that PowerExchange will search in order to qualify the first table name (if the table name is unqualified) on a select, insert, delete or update statement.

A maximum of 34 libraries can be specified, each with a maximum length of 10 characters.

Note: If both the library list override (LIBRARYLIST) and the file list override (OVRDBF) are specified, and a table exists in the LIBRARYLIST as well as in the OVRDBF, the OVRDBF will take precedence.

{DTLNOUPDATECDEP=Y|N}If set to Y, does not update the PowerExchange CDEP file for the application.

{DTLORACOLL= }Oracle CDC only. Enables the PowerExchange Listener to submit multiple simultaneous Oracle Capture processes that can connect to different Oracle instances. Oracle Capture will now use the override to determine to which instance it should connect, as opposed to using the ORACOLL keyword of the CAPI_CONNECTION TYPE=ORCL statement in dbmover.cfg.

This enables the customer to use a single PowerExchange Listener to capture data from as many as 10 Oracle instances simultaneously.

{DTLORACONN= }Oracle CDC only. Enables the user to override Oracle connection information for a given Oracle Collection Id. This means that the user can use a single set of registrations to capture data from multiple Oracle instances.

This overrides the fourth value in the ORACLEID statement of the dbmover.cfg file, for example:

ORACLEID=(coll_id,oracle_sid,connect_string,cap_connect_string)

Used in conjunction with the Oracle Instance.

Users can specify either or both Instance/Connection string. If one of the keywords is not specified, Oracle Capture will pick up the value of the other from dbmover.cfg.

{DTLORAINST= }Oracle CDC only. Enables the user to override Oracle instance information for a given Oracle Collection Id. This means that the user can use a single set of registrations to capture data from multiple Oracle instances.

This overrides the second value in the ORACLEID statement of the dbmover.cfg file, for example:

ORACLEID=(coll_id,oracle_sid,connect_string,cap_connect_string)


Used in conjunction with the Oracle Connection string.

Users can specify either or both Instance/Connection string. If one of the keywords is not specified, Oracle Capture will pick up the value of the other from dbmover.cfg.

{DTLORASCHEMA= }Oracle CDC only. Enables users to override the schema name for a group of capture registrations. This means that users can use a single set of registrations to capture data from multiple schemas that exist in a given Oracle instance.

{DTLOVRDBF= }i5/OS only. Using the database file override to replace any occurrence of the specified filename, in any SQL statement, with the new library/file/member combination. The override occurs whether the file is qualified or not and includes any files that are qualified using DTLLIBRARYLIST.

Use the following format:

from_file/to_library/to_file/to_member

- or -

from_file/to_library/to_file

Where:

♦ from_file. The file to be overridden.

♦ to_library. The new library to use.

♦ to_file. The file in the new library to use.

♦ to_member. Optional. The member in the new library and file to use. *FIRST is used if nothing is specified. In the latter case the member *FIRST will be assumed.

Note: If both the library list override (DTLLIBRARYLIST) and the database file override are specified, and a table exists in the DTLLIBRARYLIST as well as in the DTLOVRDBF, the DTLOVRDBF will take precedence.

{DTLREJECTFILE= }Overrides the asynchronous write reject file. For more information about reject files, see PowerExchange Bulk Data Movement Guide.

{DTLSESSID=}A user supplied string of maximum 8 characters. Used to link a task with statistics written to SMF or file. For more information about the statistics record layout, see “General Section” on page 81.

{DTLSTOPAFTER=}Overrides the stop on errors counter. The permitted range is 0-2147483647. For more information about the stop on errors counter override, see the PowerExchange Bulk Data Movement Guide.

{DTLSTRIPORDERBY=Y|N}If set to Y strips all following Order By clauses in the SQL. In addition, PowerExchange will treat any "--" characters in the SQL as comment indicators and will strip out the comment indicators and the following SQL text.


{DTLTIMEOUT= }Sets the maximum approximate time (in seconds) to wait for data on a queue before returning EOF.

0 indicates EOF should be returned as soon as the end of the current log is reached. The end of the log is determined at the START of the extract process (as the real end is an ever moving target).

86400 indicates that EOF is never returned. The job will wait forever.

{DTLUDBDB= }DB2 for Linux, UNIX, and Windows only. This parameter overrides the database connection obtained from the extraction map.

{DTLUPDELSEINS=Y}If set to Y, enables the update else insert mode. For more information about update else insert mode, see the PowerExchange Bulk Data Movement Guide.

{DTLWORKERS= }When you select offload processing for a PowerExchange Change Data Capture (CDC) or bulk data movement session, specifies the number of threads that PowerExchange uses on the Integration Service machine to process data. For optimal performance, the value should not exceed the number of installed or available processors on the Integration Service machine.

If you do not select offload processing, PowerExchange ignores the value of the DTLWORKERS SQL escape sequence.

Valid values are 0 to 64.

{DTLXTRASCHEMA= }This parameter overrides the schema defined in the extraction map. For more information, see “XTRASCHEMA” on page 163.

{DTLXTYPE=RS|SL}Enables users to specify the capture extract type when accessing PowerExchange Change Data Capture (CDC) through ODBC.


C H A P T E R 1 2

Datatypes and Conversion Matrix


♦ Datatype Conversions, 177

♦ ODBC Datatype Equivalency, 178

Datatype Conversions

PowerExchange supports the following conversions according to the matrix below, with a 'Y' indicating that a conversion is supported. Not all platforms support 64-bit integers. Where it is not supported PowerExchange will provide an emulation using floating point conversions. This provides precision up to approximately 15 digits, not the full 18 of a 64-bit integer.

FROM / TO

DB

L

FLT

NU

MXX

NU

MC

HA

R

DAT

E

TIM

E

TIM

ESTA

MP

PAC

KE

D/

UPA

CK

ED

ZON

ED

/U

ZON

ED

BIN

VAR

BIN

CH

AR

VAR

CH

AR

STR

ING

DBL Y Y Y Y

FLT Y Y Y Y Y

NUMxx Y Y Y Y Y Y Y Y

NUMCHAR Y Y Y Y Y Y Y Y

DATE Y Y Y Y Y

TIME Y Y Y Y Y

TIMESTAMP

Y Y Y Y Y Y

PACKED/U Y Y Y Y Y Y Y Y

ZONED/U Y Y Y Y Y Y Y Y

BIN Y Y Y Y Y

VARBIN Y Y Y Y Y

CHAR Y Y Y Y Y Y Y Y

VARCHAR Y Y Y Y Y Y Y Y

STRING Y Y Y Y Y Y Y Y

177

ODBC Datatype Equivalency

ODBC SQL Datatype ODBC SQL C Datatype PowerExchange Datatype

SQL_BINARY DTLNET_BIN

SQL_BIT SQL_C_BIT DTLNET_BIT

SQL_CHAR DTLNET_CHAR

SQL_DATE SQL_C_DATE DTLNET_DATE

SQL_DECIMAL DTLNET_MONEY

SQL_DECIMAL SQL_C_ULONG DTLNET_NUM32U

SQL_DECIMAL DTLNET_NUM64

SQL_DECIMAL DTLNET_NUM64U

SQL_DECIMAL DTLNET_NUMCHAR

SQL_DECIMAL DTLNET_PACKED

SQL_DECIMAL DTLNET_UPACKED

SQL_DECIMAL DTLNET_UZONED

SQL_DECIMAL DTLNET_ZONED

SQL_DOUBLE SQL_C_DOUBLE DTLNET_DOUBLE

SQL_INTEGER SQL_C_USHORT DTLNET_NUM16U

SQL_INTEGER SQL_C_LONG and SQL_C_SLONG

DTLNET_NUM32

SQL_REAL SQL_C_FLOAT DTLNET_FLOAT

SQL_SMALLINT SQL_C_SHORT and SQL_C_SSHORT

DTLNET_NUM16

SQL_SMALLINT SQL_C_UTINYINT DTLNET_NUM8U

SQL_TIME SQL_C_TIME DTLNET_TIME

SQL_TIMESTAMP SQL_C_TIMESTAMP DTLNET_TIMESTAMP

SQL_TINYINT SQL_C_TINYINT and SQL_C_STINYINT

DTLNET_NUM8

SQL_VARBINARY SQL_C_BINARY DTLNET_VARBIN

SQL_VARCHAR SQL_C_CHAR DTLNET_STRING

SQL_VARCHAR DTLNET_VARCHAR

178 Chapter 12: Datatypes and Conversion Matrix

A P P E N D I X A

PowerExchange Glossary

This appendix provides a glossary of terms used in PowerExchange.

Glossary of Terms

A

access method

A source-specific program interface that PowerExchange uses to extract bulk or change data from a data source, such as a table, data set, change stream, condense file, or flat file.

active logs

The disk data sets in which the PowerExchange Logger stores captured change data. See also archive logs on page 180.

after image

A representation of a source record or row after an insert or update operation. PowerExchange can save after images as part of the change data stream under certain conditions. See also before image on page 180.

Agent

See PowerExchange Agent on page 190.

APF

See authorized program facility (APF) on page 180.

application group

In the PowerExchange Navigator, a set of application names for the extraction processes that run against the data captured at a specific location. An application group is generated when you define a registration group. The application group is used to filter the data in the CDEP file for display in the PowerExchange Navigator. See also application name on page 179 and registration group on page 192.

application name

A unique name that you assign to an extraction process when you are using PowerCenter to extract the change data. In the PowerExchange Navigator, application names are shown under the application group for the location from which the change data is being extracted. See also application group on page 179.

179

Application Transparent - Transport Layer Security (AT-TLS) file

A policy file that you can define to specify the MVS jobs that are authorized to use Secure Sockets Layer (SSL) security.

archive logs

In PowerExchange, the backup data sets that the PowerExchange Logger generates on disk or tape to hold change data that is off-loaded from an active log when it becomes full. After the data is off-loaded, the logging of data to the active log can resume. See also active logs on page 179.

asynchronous capture

A type of change data capture where the changes are read from database log files instead of directly from the source database records as transactions occur. This method is used for all relational database sources and for log-based capture. See also synchronous capture on page 193.

asynchronous write

A mode of writing data to a database where the client sends blocks of data to the PowerExchange Listener and does not wait for a response before sending another block of data.

authorized program facility (APF)

An IBM facility in a z/OS environment that you can use in PowerExchange to establish authorization for programs to use restricted functions.

B

batch extraction mode

A method of extracting change data from PowerExchange Condense condense files or PowerExchange Logger for Linux, UNIX, and Windows log files, whereby the extraction runs in batch mode. Only the files that are closed when the extraction job starts are processed. After the extraction job completes reading changes from those condense files or log files, it terminates. See also continuous extraction mode on page 182 and real-time extraction mode on page 191.

Batch Message Processing (BMP)

An IMS batch processing program that has access to online databases and message queues but is started with JCL. The PowerExchange IMS synchronous ECCR can capture change data from BMP sessions.

before image

A representation of a source record or row before the last delete or update operation is processed. PowerExchange can save before images as part of the change data stream under certain conditions. See also after image on page 179.

BMP

See Batch Message Processing (BMP) on page 180.

bulk data extraction

The movement of source data in bulk to materialize a target or to synchronize a target data with the source data at a specific point in time.

C

CAPI

See consumer API (CAPI) on page 182.

180 Appendix A: PowerExchange Glossary

CAPI connection

The consumer API connection that is defined in the DBMOVER configuration file and that PowerExchange uses to extract captured change data from the change stream that is for a specific source.

CAPTPARM configuration member

The member that defines configuration parameters for PowerExchange Condense on i5/OS and z/OS. On i5/OS, this member resides in the CFG file in the datalib library and in the CFGCOND file in the condlib library. On z/OS, this member is in the RUNLIB library. Corresponds to the PWXCCL configuration file on Linux, UNIX, and Windows. See also PowerExchange Condense on page 190.

capture catalog table

The table that you create for storing information about DB2 for Linux, UNIX, and Windows tables that are registered for change data capture, including their column definitions and valid log positions. The capture catalog table must reside in the same database as the registered tables. You initialize the capture catalog table by issuing the SNAPSHOT command of the DTLUCUDB utility.

capture directory tables

A set of tables that the DB2 ECCR uses to track information about the DB2 for z/OS tables that are registered for change data capture. These tables are created at PowerExchange installation and must reside on the same DB2 subsystem as the tables registered for change data capture.

capture registration

A named definition of the source data for which you want PowerExchange to capture changes. In the PowerExchange Navigator, you define a capture registration under a registration group. The registration addresses a single table or data set and specifies the columns or fields for which to capture changes, the Condense option to use, and the registration status. See also registration group on page 192.

CAPX

An access method that PowerExchange uses to access change data in PowerExchange Condense condense files or PowerExchange Logger for Linux, UNIX, and Windows log files when you use batch extraction mode. Also, a type of CAPI_CONNECTION statement in the dbmover.cfg file for a relational database source on a Linux, UNIX, or Windows system. This statement specifies parameters for extracting change data from condense files or PowerExchange Logger log files in continuous extraction mode.

CAPXRT

An access method that the PowerExchange Listener uses to access change data from the change stream.

CCT

The file in which PowerExchange stores information about the capture registrations that users have defined in the PowerExchange Navigator.

CDC

See change data capture (CDC) on page 182 and PowerExchange Change Data Capture (CDC) on page 190.

CDCT

The file in which PowerExchange Condense on MVS or i5/OS stores information about each condense file that it generates, or the file in which the PowerExchange Logger for Linux, UNIX, and Windows stores information about each log file that it generates. This information includes the file name, start and end times, condense type, and number of records.

CDEP

The control file in which PowerExchange stores application names and information about the extraction processes that have run, including their input and timings. See also application name on page 179.

Glossary of Terms 181

Change Collector

A PowerExchange component for Datacom and IDMS sources that captures changes as they occur and passes the changes to the Change Controller. The Change Controller passes the changes to a data space that stores them until they are transmitted to the PowerExchange Logger. For Datacom sources only, you can use the more efficient direct-log-write method of capturing changes. In this case, the Change Collector passes the changes directly to the PowerExchange Logger. The Datacom Change Collector runs in the Datacom MUF address space, and the IDMS Change Collector runs in the IDMS CV address space. See also Change Controller on page 182 and Log Feeder on page 188.

Change Controller

A PowerExchange component for Datacom and IDMS sources that manages the capture registrations for the Change Collector. For IDMS sources and for any Datacom sources for which you do not use the direct-log-write method of change capture, the Change Controller also stores captured changes in an interim data space. The Log Feeder will retrieve the changes from this data space. The Change Controller runs in an address space that is separate from the Change Collector address space. See also Change Collector on page 182 and Log Feeder on page 188.

change data capture (CDC)

In PowerExchange, the process of capturing changes that were made to source databases, tables, and data sets for transmittal to a relational or nonrelational target. See also data extraction on page 183 and PowerExchange Change Data Capture (CDC) on page 190.

change stream

The change data that PowerExchange has read from the PowerExchange Logger log files on MVS systems or from RDBMS log files on other platforms.

checkpoint files

Files that PowerExchange generates to store checkpoint information for PowerExchange Condense on MVS or i5/OS or for the PowerExchange Logger on Linux, UNIX, or Windows. This information is used during a warm start to properly resume PowerExchange Condense or PowerExchange Logger processing from the point at which it was interrupted.

CmdPrefix

An MVS command prefix that you use when specifying PowerExchange Agent commands. This prefix is defined in the PowerExchange Agent AGENTCTL configuration parameters.

Condense

See PowerExchange Condense on page 190.

condense files

The files in which a PowerExchange Condense process on MVS or i5/OS stores captured change data. See also PowerExchange Condense on page 190.

consumer API (CAPI)

The API PowerExchange uses to extract the captured change data for a specific source type.

continuous extraction mode

A method of extracting change data from PowerExchange Condense condense files or PowerExchange Logger for Linux, UNIX, and Windows log files as the data is being written to the files. The extraction continues to run until you stop it. See also batch extraction mode on page 180 and real-time extraction mode on page 191.


cross-system coupling facility (XCF)

A MVS component that provides functions to support cooperation between authorized programs running within a sysplex. PowerExchange uses XCF to exchange control information between PowerExchange Change Data Capture components on a single MVS system or on multiple MVS systems in a Post-Log Merge environment.

D

data extraction

The process of extracting captured change data from the change stream so that the data can be moved to a target location. See also change data capture (CDC) on page 182 and PowerExchange Change Data Capture (CDC) on page 190.

data map

The mapping of nonrelational records to relational tables so that PowerExchange can use the SQL language for relational databases to access the data. Also, a mapping of DB2 fields so that you can split the data in a column that contains multiple fields into separate fields.

data replication

See replication on page 192.

data source

A database unit or file from which PowerExchange can capture change data or move bulk data. Sources include relational database tables, nonrelational database elements, and flat files. See also relational source on page 192 and nonrelational source on page 189.

data target

A database unit or file to which the data that PowerExchange captured can be written. PowerExchange works with PowerCenter to load captured data to a target. PowerExchange also provides methods of writing data to nonrelational targets. Targets include relational database tables, nonrelational database elements, and flat files. See also relational target on page 192 and nonrelational target on page 189.

datatype conversion

The process of converting data stored in one relational or programmatic datatype to another relational or programmatic datatype.

DBMOVER configuration file

The primary configuration file that the PowerExchange Listener uses in performing PowerExchange functions. The file includes statements that are required for PowerExchange to receive requests from remote systems for processing. It also includes statements that are required to connect to source or target databases.

Direct Log Write

The recommended method of capturing changes for Datacom sources. With this method, the Datacom Change Collector captures changes as they occur and passes them directly to the PowerExchange Logger. The capture registrations that the Change Collector uses are managed by the Change Controller. This direct-log-write method is more efficient than the older, alternative method that uses the Change Controller to store the changes in an interim data space and the Log Feeder to pass the changes from the data space to the PowerExchange Logger. See also Change Collector on page 182, Change Controller on page 182, and Log Feeder on page 188.


DTLCCADW utility

A PowerExchange utility that the Adabas ECCR runs transparently to manage the PowerExchange PCAT control file for change data processing. You can run the utility manually to override the default processing if necessary. Also called the Adabas PCAT Utility.

DTLCUIML utility

A PowerExchange utility that you can use to define a marker in the IMS logs for use by the IMS log-based ECCR. When the ECCR encounters a marker, a PowerExchange Logger message is issued that specifies restart tokens for the affected registration tags. You can use these tokens as input to the DTLUAPPL utility to define a start point for extraction processing. The DTLCUIML utility runs as an IMS Batch Message Processing (BMP) job. also called the IMS Log Marker Utility.

DTLINFO utility

A PowerExchange utility that displays the version, release, and build level for PowerExchange or for a specific PowerExchange module. Also called the Release Information Utility.

DTLMSG file

A file from which PowerExchange reads messages for various tasks and processes.

DTLREXE utility

A PowerExchange utility that you can use to submit an MVS job from a remote MVS image or from a non-MVS platform or to test the connectivity of a PowerExchange Listener. Also called Remote Execution Utility.

DTLUAPPL utility

A PowerExchange utility that sets or resets extraction restart tokens for all PowerExchange Change Data Capture sources. This utility enables you to add or edit application name entries in the Change Data Extraction Repository (CDEP) file, generate restart tokens, and print the restart tokens and application name entries. Also called the Restart Token Utility.

DTLUCBRG utility

A PowerExchange utility that you can use to create capture registrations and extraction maps in batch mode for a set of existing tables or data maps. The utility provides a way to create these items at specified PowerExchange Listener locations based on generic settings. Also called the Batch Registration Utility.

DTLUCDEP utility

A PowerExchange utility that you can use to edit or print the contents of the CDEP file. The file contains information about the extraction processes that have run, including their input and timings. When another extraction process runs, it reads the file to establish a start point. You can use the utility to remove old data from the file to prevent it from becoming too large. Also called the CDEP Utility.

DTLUCSR2 utility

A PowerExchange utility that identifies the locations of IDMS SR3 records after an event such as a database reorganization causes these records to become separated from their corresponding SR2 records. The IDMS ECCR needs the locations of the SR2 records to properly scan the SR2 and SR3 records for change data capture.

DTLUCUDB utility

A PowerExchange utility that you run shortly after installation and prior to creating capture registrations to initialize the PowerExchange capture catalog table for DB2 for Linux, UNIX, and Windows sources. You can also run the utility to generate diagnostic information for troubleshooting data capture problems. Also called DB2 CDC Utility.


DTLULCAT and DTLULOGC utilities

PowerExchange utilities for updating the IDMS Log Catalog, which is used for IDMS log-based change data capture. DTLULCAT prepares the input for DTLULOGC. DTLULOGC populates the Log Catalog with updated information about the logs to process.

DTLURDMO utility

A PowerExchange utility that you can use to copy data maps, capture registrations, and extraction maps to another location. The utility also provides a way to change the attributes of the registrations and data maps, such as the schema name or table name, during the copy process. Also called the Data Map Utility.

DTLUTSK utility

A PowerExchange utility that you can use to list all active PowerExchange tasks, locations, and allocated data sets. You can also stop tasks if necessary. Also called the Task Control Utility.

E

ECCR

See Environmental Change Capture Routine (ECCR) on page 186.

EDMC

A CICS transaction for controlling CICS/VSAM processing. This transaction is used to initialize the ECCR, terminate processing, display the files that are participating in the process, and display a help panel for the ECCR.

EDMLUCTR utility

A PowerExchange utility that you can use to display information about the PowerExchange Logger logs and the captured change data in the logs. This information includes summary information about data sources by registration tag, detail on change records and UOW records, and lists of UOWs that have not yet ended. Also called Log Scan and Print Utility.

EDMMSG data set

A SYSOUT data set on z/OS for messages from the PowerExchange Agent, ECCRs, PowerExchange Logger, Log Read API (LRAPI), and the Log Write API (LWAPI).

EDMNAME

A name that PowerExchange generates and assigns to a registered MVS source segment, table, database, or data set to identify the change records of interest for CDC processing. For MVS data sources, the terms EDMNAME and registration tag are equivalent and used interchangeably. You can view EDMNAMEs in the ECCR output in the EDMMSG SYSOUT data set. Alternatively, in the PowerExchange Navigator, look up the equivalent Tag names in capture registrations.

EDMSDIR options module

A PowerExchange module that contains the configuration options for the PowerExchange Change Data Capture components on an MVS system.

EDMUPARM options module

A PowerExchange module that contains parameters for the PowerExchange Logger on an MVS system.

EDMXLUTL utility

A PowerExchange utility for creating an event marker in the PowerExchange Logger logs on an MVS system. Also called the Event Marker Utility.


emergency restart data set (ERDS)

A bootstrap data set on an MVS system that the PowerExchange Logger uses for restarting extractions. The data set contains an inventory of the active and archive log data sets and other control information for the PowerExchange Logger.

encryption mode

The type of data encryption that PowerExchange uses when moving data. You set the encryption mode in the DBMOVER configuration file. These encryption modes are: the PowerExchange algorithm, Digital Encryption Standard (DES), Triple DES, and RC2 by RSA Security. If you use the Secure Sockets Layer (SSL) protocol to secure communications among PowerExchange systems in a network, you can also configure SSL to allow the encryption and decryption of data.

Environmental Change Capture Routine (ECCR)

A PowerExchange component that captures changes from a source database, table, or data set and writes them to log files. The log files store the changes for future extraction processing. PowerExchange Change Data Capture provides a separate ECCR for each supported source.

ERDS

See emergency restart data set (ERDS) on page 186.

error action file

In PowerExchange, a file that contains custom specifications for fault tolerance behavior in error handling. You must specify the location of customized error action files in the PowerExchange DBMOVER configuration file on the target data platform.

event marker

In the PowerExchange Logger, a record that represents a specific point in the change stream. Event markers can be used to provide restart points for extractions.

extended recovery facility (XRF)

An IMS facility for minimizing the effects of failures or stoppages in z/OS, VTAM, the host processor, or high-availability applications during sessions between high-availability applications and designated terminals. This facility provides an alternative IMS subsystem that can take over sessions from the failing subsystem.

extraction

See data extraction on page 183.

extraction group

In the PowerExchange Navigator, a set of extraction maps for extracting the data that is captured from a specific database instance and node location using a specific access method. An extraction group is generated when you define a registration group. The extraction group is used to filter the data in the DTLCAMAP file (on MVS systems) or the CAMAPS file (on non-MVS systems) for display in the PowerExchange Navigator. See also extraction map on page 186 and registration group on page 192.

extraction map

A definition of a change data extraction for a source table or data set. PowerExchange generates an extraction map for each capture registration that you define. You can edit the generated extraction maps to add or remove columns. You can also create additional extraction maps for a capture registration. In the PowerExchange Navigator, extraction maps are shown under the extraction group for the database instance and node location from which the data will be extracted. See also extraction group on page 186.


extraction mode

A method of extracting change data from the change stream. See also batch extraction mode on page 180, continuous extraction mode on page 182, and real-time extraction mode on page 191.

F

Flexible Target Key Custom Transformation

A type of transformation that you can define in PowerCenter for a PowerExchange Change Data Capture source. This transformation is useful when the source table contains key columns that are subject to UPDATE SQL commands and the target has key columns that are mapped to those source columns. This transformation generates a pair of delete and insert statements to update the target instead of using the update statement.

full condense

A condense type that is available for PowerExchange Condense on MVS or i5/OS. With this condense type, change data is accumulated in keyed condense files such that later changes supersede earlier changes. For example, if insert1, update1, and update2 are written to a source field, in that sequence, each change is passed to the condense file but only update2 is available for extraction processing. To use this condense type, you must select Full for the Condense option when defining capture registrations in the PowerExchange Navigator. This condense type does not maintain transactional consistency. See also partial condense on page 190.

G

group definition file

See PowerExchange Condense group definition file on page 190 or PowerExchange Logger group definition file on page 191.

group source

When using PWXPC connections in PowerCenter workflows, the manner in which PowerExchange reads all data from a physical source in a single pass. For change data, PowerExchange reads the data in the change stream once for all tables defined in the mapping. For multiple-record VSAM and sequential data sets, PowerExchange reads the data set once to acquire all record types.

H

HOSTENT utility

A PowerExchange utility that you can use to display the TCP/IP host name and address for a system and to diagnose problems related to PowerExchange communication and licensing. Also called TCP/IP Address Reporter utility.

I

ICU

See International Component for Unicode (ICU) on page 187.

in-doubt UOWs

The units of work that have not yet been committed to the source database.

International Component for Unicode (ICU)

A portable set of C/C++ and Java libraries for Unicode support, software internationalization (I18N), and globalization (G11N) code page support, which is provided with PowerExchange.


L

listener

See PowerExchange Listener on page 191.

local mode

A method that a process, such as an extraction process, uses to communicate with PowerExchange when the process and PowerExchange are running on the same system. This method does not use IP.

location

In PowerExchange, a user-specified node statement in the DBMOVER configuration file that identifies a PowerExchange Listener on a remote platform. The node statement contains a port number and the IP address or host name of the remote source or target platform.

log-based capture

An asynchronous method of reading change data from log files.

Log Feeder

A PowerExchange component for Datacom and IDMS sources that reads captured changes from the data space in which the Change Controller stores them and passes the changes to PowerExchange Logger log files. The Log Feeder runs an address space that is separate from those of the other change-data-capture components. If you use the direct-log-write method of capturing changes for Datacom sources, the Log Feeder is not used for change data capture. See also Change Collector on page 182 and Change Controller on page 182.

Logger

See PowerExchange Logger for Linux, UNIX, and Windows on page 191 and PowerExchange Logger for MVS on page 191.

logger token

See restart tokens on page 192.

Log Read API (LRAPI)

A PowerExchange API that connects to the PowerExchange Logger to read the captured change data from the log files.

Log Write API (LWAPI)

A PowerExchange API that connects to the PowerExchange Logger to write the change data that was captured by the ECCR to the log files.

LRAPI

See Log Read API (LRAPI) on page 188.

LWAPI

See Log Write API (LWAPI) on page 188.

M

makeconv utility

An open source program provided by the International Component for Unicode (ICU) that converts a source code page definition from a Unified Change Management (UCM) file into a binary CNV file.


materialization

The initial loading of bulk data to a target. After the data loaded, the target is materialized. You can then propagate only the change data to update it.

message destination override

A way for users to suppress PowerExchange messages or to direct these messages to specific locations.

MUFplex

For Datacom sources, a collection of Multi-User Facility (MUF) systems that can share data.

N

Navigator

See PowerExchange Navigator on page 191.

near real time

A mode in which PowerExchange provides change data as the changes are made.

Netport job

A JCL stream that you can associate with a port to accommodate special situations. For example, you can use a Netport job to access sources through a non-PowerExchange module such as IMS DLIBATCH, alleviate excessive wait times for tape mounts, perform GDG processing, or provide IDMS access with full user ID checking. You link a Netport job to a port by setting parameters in the DBMOVER configuration file.

node

In PowerExchange, a remote PowerExchange Listener to which you can connect.

nonrelational source

In PowerExchange, a unit in a database other than a relational database from which PowerExchange can capture changes or move bulk data. Nonrelational sources include IMS segments, IDMS records, Adabas files, VSAM data sets, and flat files. See also relational target on page 192.

nonrelational target

In PowerExchange, a unit in a database other than a relational database to which captured changes or bulk data can be written. Nonrelational targets include IMS segments, IDMS records, Adabas files, VSAM data sets, and flat files. See also relational source on page 192.

normalization

The process of restructuring a data model by reducing its relations to their simplest forms when designing a logical relational database. Normalization helps avoid redundancies and inconsistencies in data.

O

ODBC

See Open Database Connectivity (ODBC) on page 189.

Open Database Connectivity (ODBC)

A standard API that a database application can use to access data in relational and nonrelational database management systems (DBMSs) that use different data storage formats and programming interfaces. PowerExchange provides a thin ODBC driver to connect with PowerCenter so that PowerCenter can read the captured data and write it to various targets. However, to integrate PowerExchange and PowerCenter, it is


recommended that you use the PowerExchange Client for PowerCenter (PWXPC) instead of the ODBC driver. See also PowerExchange Client for PowerCenter (PWXPC) on page 190.

P

partial condense

A condense type that is available for PowerExchange Condense or for the PowerExchange Logger for Linux, UNIX, and Windows. With this condense type, change data is distilled and written to either PowerExchange Condense condense files or PowerExchange Logger log files, in chronological order by UOW end time. To use partial condense, you must select Part for the Condense option when defining capture registrations in the PowerExchange Navigator. This condense type maintains transactional consistency. See also full condense on page 187.

personal metadata

The parameters that you define in the PowerExchange Navigator for accessing metadata about the bulk and change data sources that you are interested in. You can also use this information to view data from those sources. These parameters are saved to a personal profile for reuse.

Post-Log Merge

A PowerExchange configuration that uses a job for extracting data that was logged by multiple PowerExchange Loggers. The logged data is merged chronologically for extraction processing. This configuration accommodates sites that use multiple MVS images and multiple PowerExchange Loggers.

PowerExchange Agent

A PowerExchange component that provides services to other PowerExchange components, including obtaining and managing global queues, creating address spaces, and managing capture registrations.

PowerExchange Call Level Interface (SCLI)

An interface that PowerExchange uses to read and write data. Data is read using SQL SELECT statements. Data is written using SQL INSERT statements.

PowerExchange Change Data Capture (CDC)

The PowerExchange feature for capturing changes to data in a source database.

PowerExchange client

PowerExchange software that enables you to use PowerExchange from a remote computer. The client communicates with a PowerExchange Listener on a different computer.

PowerExchange Client for PowerCenter (PWXPC)

A PowerCenter component that fully integrates PowerCenter with PowerExchange so that PowerCenter can access PowerExchange-controlled data and write it to various targets. PWXPC has several advantages over the alternative ODBC drivers, which PowerExchange also provides for this purpose.

PowerExchange Condense

An optional PowerExchange process for data sources on MVS or i5/OS. PowerExchange Condense condenses captured change data and stores the data in condense files. The condense files store only the successfully completed UOWs, in chronological order by end time. When an extraction session runs, the data is extracted from the condense files rather than from the change stream.

PowerExchange Condense group definition file

An optional file that defines a group of capture registrations that PowerExchange Condense uses to write change data for registered tables to separate condense files. The use of a separate condense file for each table in a group


makes extraction processing of that data more efficient. The group definition file name must be defined in the GROUPDEFS parameter of the dtlca.cfg file. See also PowerExchange Condense on page 190.

PowerExchange Listener

The PowerExchange feature that coordinates activities across platforms, initiates requests, or processes requests from third-party applications. A PowerExchange Listener must be started if you plan to move data across platforms.

PowerExchange Logger for Linux, UNIX, and Windows

An optional PowerExchange process for data sources on Linux, UNIX, and Windows. The PowerExchange Logger captures and stores change data in PowerExchange Logger log files. The log files store only the successfully committed UOWs, in chronological order by end time. When an extraction session runs, the change data is extracted from the PowerExchange Logger log files rather than from the database logs.

PowerExchange Logger for MVS

A PowerExchange component that is required to perform change data capture for data sources on MVS. The PowerExchange Logger receives captured change data from an ECCR, creates a persistent copy of the data, and inventories the data. When an extraction session runs, the PowerExchange Logger supplies the data on demand for replication to target databases. See also Environmental Change Capture Routine (ECCR) on page 186.

PowerExchange Logger group definition file

An optional file that defines a group of capture registrations that the PowerExchange Logger for Linux, UNIX, and Windows uses to write change data for registered tables to separate log files. The use of a separate log file for each table in a group makes extraction processing of that data more efficient. The file name must be defined in the GROUPDEFS parameter of the pwxccl.cfg file. See also PowerExchange Logger for Linux, UNIX, and Windows on page 191.

PowerExchange Navigator

A PowerExchange Windows client for defining and managing change data and bulk data movements. You must use the PowerExchange Navigator to create capture registrations for data sources. You can also it to create and edit extraction maps, data maps, and personal metadata profiles.

PWXCCL configuration file

The file that defines parameters for configuring the PowerExchange Logger for Linux, UNIX, and Windows. Corresponds to the CAPTPARM member on i5/OS and z/OS. See also PowerExchange Logger for Linux, UNIX, and Windows on page 191.

PWXPC

See PowerExchange Client for PowerCenter (PWXPC) on page 190.

R

real time

In PowerExchange, a change data capture method whereby PowerExchange retrieves the data changes as they are made to a source database.

real-time extraction mode

In PowerExchange, a method of extracting change data from the change stream in near real time, as the changes are made. The extraction continues to run until you stop it. See also batch extraction mode on page 180 and continuous extraction mode on page 182.


register

In PowerExchange, to define the source data resources that you want to participate in data propagation. These definitions are saved to the PowerExchange CCT file. PowerExchange recognizes only the source data that you register. See also capture registration on page 181.

registration group

In the PowerExchange Navigator, a named set of capture registrations for capturing source data. The definition includes the PowerExchange Listener location, the database type, and the user ID and password for accessing the source data. The registration group is used to filter the data in the CCT file for display in the PowerExchange Navigator. When you define a registration group, a corresponding application group and extraction group are generated. See also capture registration on page 181, application group on page 179, and extraction group on page 186.

registration tag

A name that PowerExchange generates and assigns to a registered source segment, table, database, or data set to identify the change records of interest for CDC processing. You can view tag names from the PowerExchange Navigator by opening capture registrations. See also EDMNAME on page 185.

reject file

A PowerExchange file that contains rows of data that are rejected and are not written to the target database or flat file. PowerExchange reject files are created on the target platform for asynchronous write operations.

relational source

In PowerExchange, a table in a relational database, such as DB2 or Oracle, from which PowerExchange can capture changes or move bulk data. See also nonrelational source on page 189.

relational target

In PowerExchange, a table in a relational database to which captured changes or bulk data can be written. See also nonrelational target on page 189.

replication

The process of maintaining the same data in multiple locations. PowerExchange in conjunction with PowerCenter replicates data by copying data from a source database and applying that data to a target database, according to your specifications.

repository

In PowerExchange, a collection of all capture registrations for z/OS data sources that are managed by a specific PowerExchange Agent.

restart tokens

Identifiers that determine the start point in the change stream for a PowerExchange extraction. The restart tokens consist of a sequence token (RESTART1) and a logger token (RESTART2). The format and content of these tokens vary by platform and source type.

row test

The process of testing a data map, extraction map, or personal metadata profile in the PowerExchange Navigator. The test accesses columns in a data source and displays them in tabular format. The results indicate the change data that PowerExchange will retrieve at runtime or the changes that were retrieved.


S

schema

A description of the fields or columns in a particular segment, record, or table. This description includes characteristics such as data type, length, and precision.

SCLI

See PowerExchange Call Level Interface (SCLI) on page 190.

sequence token

See restart tokens on page 192.

Shadow MUF

For Datacom sources, a Multi-User Facility system within the MUFplex that can take over processing when its partner MUF system fails. See also MUFplex on page 189.

source

See data source on page 183.

SQL escape sequence

A method for overriding the PowerExchange ODBC settings during execution.

synchronization

In a data replication environment, the process of ensuring, at a single point in time, that a set of data in a target database is an accurate representation of the corresponding set of data in the source database. By moving change data in real time, PowerExchange helps to achieve data synchronization.

synchronous capture

A type of change data capture where the PowerExchange ECCR captures changes from a source as they are made and sends them directly to the PowerExchange Logger logs, where they are available for extraction. This type of capture is used for Datacom, IDMS synchronous, IMS synchronous, batch VSAM, and CICS/VSAM sources. IMS and IDMS log-based sources use asynchronous capture. See also asynchronous capture on page 180.

synchronous write

A mode of writing data to a database, table, or file in real time using PowerExchange, where a record or row must be successfully written before the next one is written. See also asynchronous write on page 180.

T

tag

See registration tag on page 192.

target

See data target on page 183.

U

unit of work (UOW)

A recoverable sequence of operations that is performed by an application between two points of consistency. A unit of work begins when a transaction starts or at a user-requested syncpoint, and it ends when the transaction ends or at a user-requested syncpoint.


UOW

See unit of work (UOW) on page 193.

UOWC

An acronym for the UOW Cleanser and also a type of CAPI connection statement. You can define a CAPI UOWC connection statement in the DBMOVER configuration file to specify parameters for the UOW Cleanser when capturing changes from MVS, DB2 for i5/OS, and Oracle sources. See also UOW Cleanser on page 194.

UOW Cleanser

A PowerExchange component that reconstructs units of work (UOWs) from a change stream into complete and consecutive UOWs that are in chronological order by end time.

UOW spill file

A file in which the UOW Cleanser stores changes for a UOW that it is processing when memory cache is insufficient to hold the entire UOW.

X

XCF

See cross-system coupling facility (XCF) on page 183.

XRF

See extended recovery facility (XRF) on page 186.


I N D E X

AABEND_SW parameter 4Access method

SMF Field 83accessing

multibyte metadata with ODBC 170ADA_L3_ALLOW parameter 6Adabas

security 98Adabas L3 logical reads 6Adabas optimization

configuration parameter 6ADABAS_DEFAULT_DBID parameter 5ADABAS_PREFIX parameter 6ADABASCODEPAGE parameter 5ADAOPT parameter 6ADAOPTM parameter 6ADAPREFETCH parameter 6ADASTATS parameter 6Additional Info

SMF Field 81AGENT 51alternative log format 36AM Type

SMF Field 83APF

SMF Requirement 78APPBUFSIZE parameter 7AS/400 Properties tab 167AS400 Extracts

Restarting ignoring in-flight UOW 49AS4J CAPI 45AS4JRNEXIT 46Authorization id

SMF Field 83

BBS parameter 8Bytes Read

SMF Field 83Bytes received

SMF Field 82Bytes sent

SMF Field 82Bytes Written

SMF Field 83

CCAPI connection statements

AS4J parameters 45CAPX CAPI_CONNECTION statement 49LRAP parameters 50MSQL CAPI_CONNECTION statement 51ORCL CAPI_CONNECTION statement 53overview 42UDB CAPI_CONNECTION statement 58UOWC CAPI_CONNECTION statement 60

CAPI_CONN_NAME parameter 8CAPI_CONNECTION parameter 8CAPI_SRC_DFLT parameter 8CAPT_PATH parameter 9CAPT_XTRA parameter 9capture registrations

security 95Capture_connect_string 27CAPX CAPI_CONNECTION statement

DFLTINST parameter 49DLLTRACE parameter 49FILEWAIT parameter 50NAME parameter 50parameters and syntax 49RSTRADV parameter 50TRACE parameter 50TYPE parameter 50

CAPX Properties tab 168CAPXRT Properties tab 168Client and Server Request Certificate Authentication 108Client IP addr

SMF Field 81Client Session ID

SMF Field 81Client UserID

SMF Field 81Clock time (end)

SMF Parameter 79Clock time (start)

SMF Field 79CMDNODE parameter 9, 65CODEPAGE parameter 10Collection_id 27COLON parameter 10COMMA parameter 10Component

SMF Field 81

195

COMPRESS parameter 10configuration parameter

ADAOPT 6NRDB_WRITE_CHAR_NULL_FILL 25NRDB_WRITE_NUM_NULL_FILL 25

Configuring for SMFDBMOVER Parameters 78

Connect_string 27Connection Type

SMF Field 83CONSOLE_CODEPAGE parameter 9CONSOLE_TRACE parameter 10CONVCHAR parameter 11conversion matrix

datatypes 177conversions

datatype 177Correlation id

SMF Field 83CPU Time

SMF Field 81CPU time

SMF Field 82CPX_DIR parameter 11creating an ODBC data source

UNIX 169Windows 164

CREDENTIALS_CASE parameter 11

Ddata map

security 96Datacom

security 99DATAMAP_SERVER parameter 12datatype

conversion matrix 177conversions 177ODBC equivalency 178

DATERANGE parameter 12DB2 for z/OS

security 99DB2 Plan Name

SMF Field 83DB2 Time

SMF Field 83DB2/S390 Bulk Load Properties tab 167DB2_BIN_AS_CHAR parameter 12DB2_BIN_CODEPAGE

AS400 codepage parameter 13DB2_ERRORFILE parameter 13DB2CODEPAGE parameter 14DB2DEF_ENCODING parameter 15DB2ID parameter 15DB2PLAN parameter 15DBMOVER configuration file

CAPI connection statements 42examples 66ICU parameters 62overview 3parameter descriptions 4

SECURITY statement 86syntax rules 4

DBMOVER configuration parameterADA_L3_ALLOW 6

DBQUAL1 159DECPOINT parameter 16DEFAULTCHAR parameter 16DEFAULTDATE parameter 16Deleted

SMF Field 83Descriptors length

SMF Field 81DISP parameter 16DM_RESOURCE parameter 16DM_SUBTASK parameter 17DMX_DIR parameter 16DTLDESCRIBE

overview 113qualifiers 129supported syntax 113

DTLMSG_CODEPAGE parameter 17

EEMR Properties tab 167ENCRYPT parameter 17ENCRYPTLEVEL parameter 18End time

SMF Field 82End Time2

SMF Field 81ENQMAJORNAME parameter 18ENQSYSTEMS parameter 18EOF 47, 51ERRROWNOTFOUND parameter 18EXT_CP_SUPPT parameter 18extraction maps

security 95Eye catcher

SMF Parameter 79

FFACILITY class 17

GGDGLOCATE 71GDGLOCATE parameter 18General tab 166

Ii5/OS

security 87

196 Index

ICU parameters 62ICUALIAS parameter 18, 62ICUCNVPROPERTY parameter 18, 62ICUCONVERTER parameter 18, 64ICUDATADIR parameter 18, 64IDMS

Netport Job Requirement 68Netport Jobs 67

IMSNetport Jobs 67security 100

IMSID parameter 19Inserted

SMF Field 83INST 47International Components for Unicode (ICU) 131Ip Address

SMF Field 81

JJob Name

SMF Field 81Job/STC name

SMF Field 81JOBCLASS parameter 19JOURNAL 47

LLength of section

SMF Field 80LIBASUSER 47Linux

security 89LISTENER parameter 20LOADCTLFILE parameter 20LOADJOBFILE parameter 20LOG 51LOG_CODEPAGE parameter 20LOGNODE parameter 20LOGPATH parameter 21LOGSID parameter 21LRAP CAPI_CONNECTION parameters 50LRECL parameter 21

MMAXTASKS parameter 21Messages received

SMF Field 82Messages sent

SMF Field 82metadata syntax 117MQAUTOCREATE parameter 21MQRECOVER parameter 21MQRESTART parameter 22MSGPREFIX parameter 22MSGPREFIX-HYPHEN parameter 22MSQL CAPI_CONNECTION statement

DISTDB parameter 52

DISTSRV parameter 52DLLTRACE parameter 52DWFLAGS parameter 52MEMCACHE parameter 52NAME parameter 52parameters and syntax 51POLWAIT parameter 52RSTRADV parameter 52TRACE parameter 53TYPE parameter 53

MVSsecurity 91

MVSDB2AF parameter 22

NNAME 47, 51NEGSIGN parameter 22Netport Jobs

Configuration 67Creating 68IDMS 67Reasons for using 67

NETPORT parameter 23No of Extended Section

SMF Parameter 79No. Rows

SMF Field 83Node name

SMF Field 81NODE parameter 24Nodename

SMF Field 81NOGETHOSTBYNAME parameter 24nonrelational SQL syntax 113NRDB_WRITE_CHAR_NULL_FILL parameter 25NRDB_WRITE_NUM_NULL_FILL parameter 25null fields

Set To Specified Value 25Number accepted connections

SMF Field 82Number connections

SMF Field 82NUMERICSIGN parameter 25

OODBASUPP parameter 25ODBC

AS/400 Properties tab 167CAPX Properties tab 168CAPXRT Properties tab 168DB2/S390 Bulk Load Properties tab 167EMR Properties tab 167General tab 166PowerExchange Data Source tab 165

ODBC datatype equivalency 178offload processing 176

Restrictions 25OFFLOADPROCESSING parameter 25Offset to 1st Extended Section

SMF Parameter 79

Index 197

ORA_ERRORFILE parameter 25Oracle_SID 27ORACLECODEPAGE 136ORACLECODEPAGE parameter 26ORACLEDB parameter 26ORACLEID parameter 27ORCL CAPI_CONNECTION statement

ARRAYSIZE parameter 53BYPASSUF parameter 54CATBEGIN parameter 54CATEND parameter 54CATINT parameter 54COMMITINT parameter 54DLLTRACE parameter 54GENRLOCK parameter 55IGNUFMSG parameter 55LGTHREAD parameter 56LOGDEST parameter 56NAME parameter 56ONLINECAT parameter 57ORACOLL parameter 56parameters and syntax 53SELRETRY parameter 57SNGLINST parameter 58TRACE parameter 58TYPE parameter 58

OUSP parameter 27

PPC_AUTH parameter 28PIPE parameter 28POLLTIME parameter 28POLWAIT 48Port Number

SMF Field 82PowerExchange Data Source tab 165PowerExchange ODBC driver 157PowerExchange ODBC driver parameters

DB2 for z/OS bulk 163general 157overview 157

PowerExchange selective sign-onsecurity 100

PowerExchange sign-on filesecurity 101

PRE861_COMPAT parameter 28PRGIND parameter 28PRGINT parameter 29Process ID

SMF Field 81Product

SMF Field 81pwxcmd program

security 90PWXHDR _

SMF Eye catcher 79, 80

PWXSOMAXCONN parameter 29

RRACF_CLASS parameter 17, 29RC 1

SMF Field 82RC 2

SMF Field 82RC 3

SMF Field 82RDBMSINSRTDFLT parameter 29Reason Code

SMF Field 81, 83RECFM parameter 29REJECT_FILE_DELIMITER parameter 29RELEASE parameter 29Request Client Authentication 108Request Server Authentication 108Return Code

SMF Field 81Rows Read

SMF Field 83Rows Written

SMF Field 83

SSection Count

SMF Field 80Section type

SMF Field 80security

Adabas 98capture registrations 95data maps 96Datacom table-level 99Datacom, table-level 99DB2 for z/OS 99extraction maps 95i5/OS 87IMS 100Linux 89MVS 91PowerExchange selective sign-on 100PowerExchange sign-on file 101pwxcmd program 90UNIX 89Windows 89z/OS 91

SECURITY parameter 31SECURITY statement

DBMOVER configuration file 86selective sign-on 104

security 100SHOW_THREAD_PERF parameter 31sign-on file

security 101SMF

APF authorisation requirement 78Extended Section 81Extended Section Descriptor 80

198 Index

Extended Section for Data Access 83Extended Section for PowerExchange Client 82General Section 81Header 78Interval 78PowerExchange Header/Descriptor 79Record Number 78Standard SMF Header with Sub Type included 79STATS DBMOVER parameter 78Sub Type 78

SMF Record Format 78SMF Recording

Overview 77SMF Statistics Logging 77SMFxDTE

SMF Field 79SMFxFLG

SMF Field 79SMFxLEN

SMF Field 79SMFxRTY

SMF Field 79SMFxSEG

SMF Field 79SMFxSID

SMF Field 79SMFxSSI

SMF Field 79SMFxSTY

SMF Parameter 79SMFxTME

SMF Field 79SPACE parameter 32SQL Code

SMF Field 83SQL escape sequences

DTLADAPREFETCH 172DTLADAPWD 172DTLAPP 172DTLCONFWRITE 172DTLCONNOVR 172DTLDATAPWD 172DTLDB2DEGREE 172DTLDSN 173DTLEVENTTB 173DTLIMGOV 173DTLIMTYPE 173DTLINSMODE 174DTLJRNL 174DTLLIBRARYLIST 174DTLNOUPDATECDEP 174DTLORACOLL 174DTLORACONN 174DTLORAINST 174DTLORASCHEMA 175DTLOVRDBF 175DTLREJECTFILE 175DTLSESSID 175DTLSTOPAFTER 175DTLSTRIPORDERBY 175DTLTIMEOUT 176DTLUDBDB 176DTLUPDELSEINS 176

DTLWORKERS 176DTLXTRASCHEMA 176DTLXTYPE 176ODBC 171with PowerCenter 171

SSIDSMF Field 83

SSL Authentication Mode 108SSL parameter 32SSL_ALLOW_SELFSIGNED parameter 32SSL_REQ_CLNT_CERT parameter 32SSL_REQ_SRVR_CERT parameter 32Start Time

SMF Field 81Start time

SMF Field 82Statistics

configuring for UDB on Windows 78Filename 78STATS DBMOVER parameter 78

statisticsInterval 78

STATSFILE DBMOVER parameter 78SMF DBMOVER parameter 78

STATS parameter 32STCK end

SMF Field 82STCK start

SMF Field 82STOPIT 48Sub type

SMF Field 81subdescriptors

configuration parameter ADAOPT 6SUBMITTIMEOUT parameter 32superdescriptors

configuration parameter ADAOPT 6SUPPRESS_DATA_LOGGING parameter 33SVCNODE parameter 33, 65SYSOUT_TIMESTAMP parameter 33

TTAPEWAIT parameter 33TCPIP_ASYNC parameter 33TCPIPBUFSIZE parameter 33TCPIPVER parameter 33TEMPHLQ parameter 34TEXT_EOF_FOR_BINARY parameter 34TIMEOUTS parameter 34TRACE parameter 34TRACING parameter 34TYPE 49, 51

UUDB CAPI_CONNECTION statement

CCATALOG parameter 58DBCONN parameter 58DLLTRACE parameter 58EPWD parameter 59

Index 199

MEMCACHE parameter 59NAME parameter 59parameters and syntax 58PASSWORD parameter 59RSTRADV parameter 59TRACE parameter 59TYPE parameter 59UDBSCHEMA parameter 59UPDINT parameter 60UPDREC parameter 60USERID parameter 60

UIDFMT 51UNIT parameter 34UNIX

security 89UOWC CAPI_CONNECTION statement

BLKSIZE parameter 60CAPINAME parameter 60DATACLASS parameter 60DLLTRACE parameter 61MEMCACHE parameter 61NAME parameter 61parameters and syntax 60RSTRADV parameter 61SPACEPRI parameter 61SPACETYP parameter 61STORCLASS parameter 61TRACE parameter 61TYPE parameter 62UNIT parameter 62

UOWRSTANYCAPI AS400 Parameter 45, 49

UpdatedSMF Field 83

VVersion/Build

SMF Field 81VOLSER parameter 34VSAM parameter 35

WWAITDSN parameter 36Windows

security 89

Zz/OS

security 91

200 Index

NOTICES

This Informatica product (the “Software”) includes certain drivers (the “DataDirect Drivers”) from DataDirect Technologies, an operating company of Progress Software Corporation (“DataDirect”) which are subject to the following terms and conditions:

1. THE DATADIRECT DRIVERS ARE PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.

2. IN NO EVENT WILL DATADIRECT OR ITS THIRD PARTY SUPPLIERS BE LIABLE TO THE END-USER CUSTOMER FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL OR OTHER DAMAGES ARISING OUT OF THE USE OF THE ODBC DRIVERS, WHETHER OR NOT INFORMED OF THE POSSIBILITIES OF DAMAGES IN ADVANCE. THESE LIMITATIONS APPLY TO ALL CAUSES OF ACTION, INCLUDING, WITHOUT LIMITATION, BREACH OF CONTRACT, BREACH OF WARRANTY, NEGLIGENCE, STRICT LIABILITY, MISREPRESENTATION AND OTHER TORTS.

Documents

PowerExchange Reference Manual - Informatica · The site contains product information, user group information, newsletters, access to the Informatica customer support case management