System Administration Guide - Informatica · Informatica Data Explorer Standard Edition System...

System Administration Guide

Informatica Data Explorer Standard Edition(Version 9.0.1)

Informatica Data Explorer Standard Edition System Administration GuideVersion 9.0.1

June 2010

Copyright © 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© (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, Informatica Platform, Informatica Data Services, PowerCenter, PowerCenterRT, PowerCenter Connect, PowerCenter Data Analyzer, PowerExchange, PowerMart, Metadata Manager, Informatica Data Quality, Informatica Data Explorer, Informatica B2B Data Transformation, 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.

Portions of this software and/or documentation are subject to copyright held by third parties, including without limitation: Copyright DataDirect Technologies. All rights reserved. Copyright © Sun Microsystems. All rights reserved. Copyright © RSA Security Inc. All Rights Reserved. Copyright © Ordinal Technology Corp. All rights reserved. Copyright © Aandacht c.v. All rights reserved. Copyright Genivia, Inc. All rights reserved. Copyright 2007 Isomorphic Software. All rights reserved. Copyright © Meta Integration Technology, Inc. All rights reserved. Copyright © Intalio. All rights reserved. Copyright © Oracle. All rights reserved. Copyright © Adobe Systems Incorporated. All rights reserved. Copyright © DataArt, Inc. All rights reserved. Copyright © ComponentSource. All rights reserved. Copyright © Microsoft Corporation. All rights reserved. Copyright © Rouge Wave Software, Inc. All rights reserved. Copyright © Teradata Corporation. All rights reserved. Copyright © Yahoo! Inc. All rights reserved. Copyright © Glyph & Cog, LLC. All rights reserved.

This product includes software developed by the Apache Software Foundation (http://www.apache.org/), and other software which is licensed under the Apache License, Version 2.0 (the "License"). You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0. Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

This product includes software which was developed by Mozilla (http://www.mozilla.org/), software copyright The JBoss Group, LLC, all rights reserved; software copyright © 1999-2006 by Bruno Lowagie and Paulo Soares and other software which is licensed under the GNU Lesser General Public License Agreement, which may be found at http://www.gnu.org/licenses/lgpl.html. The materials are provided free of charge by Informatica, "as-is", without warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability and fitness for a particular purpose.

The product includes ACE(TM) and TAO(TM) software copyrighted by Douglas C. Schmidt and his research group at Washington University, University of California, Irvine, and Vanderbilt University, Copyright © 1993-2006, all rights reserved.

This product includes software developed by the OpenSSL Project for use in the OpenSSL Toolkit (copyright The OpenSSL Project. All Rights Reserved) and redistribution of this software is subject to terms available at http://www.openssl.org.

This product includes Curl software which is Copyright 1996-2007, Daniel Stenberg, <[email protected]>. All Rights Reserved. Permissions and limitations regarding this software are subject to terms available at http://curl.haxx.se/docs/copyright.html. Permission to use, copy, modify, and distribute this software for any purpose with or without fee is hereby granted, provided that the above copyright notice and this permission notice appear in all copies.

The product includes software copyright 2001-2005 © MetaStuff, Ltd. All Rights Reserved. Permissions and limitations regarding this software are subject to terms available at http://www.dom4j.org/ license.html.

The product includes software copyright © 2004-2007, The Dojo Foundation. All Rights Reserved. Permissions and limitations regarding this software are subject to terms available at http:// svn.dojotoolkit.org/dojo/trunk/LICENSE.

This product includes ICU software which is copyright International Business Machines Corporation and others. All rights reserved. Permissions and limitations regarding this software are subject to terms available at http://source.icu-project.org/repos/icu/icu/trunk/license.html.

This product includes software copyright © 1996-2006 Per Bothner. All rights reserved. Your right to use such materials is set forth in the license which may be found at http://www.gnu.org/software/ kawa/Software-License.html.

This product includes OSSP UUID software which is Copyright © 2002 Ralf S. Engelschall, Copyright © 2002 The OSSP Project Copyright © 2002 Cable & Wireless Deutschland. Permissions and limitations regarding this software are subject to terms available at http://www.opensource.org/licenses/mit-license.php.

This product includes software developed by Boost (http://www.boost.org/) or under the Boost software license. Permissions and limitations regarding this software are subject to terms available at http:/ /www.boost.org/LICENSE_1_0.txt.

This product includes software copyright © 1997-2007 University of Cambridge. Permissions and limitations regarding this software are subject to terms available at http://www.pcre.org/license.txt.

This product includes software copyright © 2007 The Eclipse Foundation. All Rights Reserved. Permissions and limitations regarding this software are subject to terms available at http:// www.eclipse.org/org/documents/epl-v10.php.

This product includes software licensed under the terms at http://www.tcl.tk/software/tcltk/license.html, http://www.bosrup.com/web/overlib/?License, http://www.stlport.org/doc/license.html, http://www.asm.ow2.org/license.html, http://www.cryptix.org/LICENSE.TXT, http://hsqldb.org/web/hsqlLicense.html, http://httpunit.sourceforge.net/doc/license.html, http://jung.sourceforge.net/license.txt , http://www.gzip.org/zlib/zlib_license.html, and http://www.openldap.org/software/release/license.html, http://www.libssh2.org, http://slf4j.org/license.html., and http://www.sente.ch/software/OpenSourceLicense.html

This product includes software licensed under the Academic Free License (http://www.opensource.org/licenses/afl-3.0.php), the Common Development and Distribution License (http://www.opensource.org/licenses/cddl1.php) the Common Public License (http://www.opensource.org/licenses/cpl1.0.php) and the BSD License (http://www.opensource.org/licenses/bsd-license.php).

This product includes software copyright © 2003-2006 Joe WaInes, 2006-2007 XStream Committers. All rights reserved. Permissions and limitations regarding this software are subject to terms available at http://xstream.codehaus.org/license.html. This product includes software developed by the Indiana University Extreme! Lab. For further information please visit http://www.extreme.indiana.edu/.

This Software is protected by U.S. Patent Numbers 5,794,246; 6,014,670; 6,016,501; 6,029,178; 6,032,158; 6,035,307; 6,044,374; 6,092,086; 6,208,990; 6,339,775; 6,640,226; 6,789,096; 6,820,077; 6,823,373; 6,850,947; 6,895,471; 7,117,215; 7,162,643; 7,254,590; 7,281,001; 7,421,458; and 7,584,422, international Patents and other Patents Pending.

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.

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.

Part Number: IDE-SAG-90100-0001

Table of Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Informatica Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica Customer Portal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica How-To Library . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica Knowledge Base . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

Informatica Multimedia Knowledge Base . . . . . . . . . . . . . . . . . . . . . . . xiii

Informatica Global Customer Support . . . . . . . . . . . . . . . . . . . . . . . . . xiii

Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Data Explorer Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Data Explorer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Data Explorer Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Data Explorer Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Data Explorer Importers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Data Explorer Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Data Explorer Import for VSAM and IMS . . . . . . . . . . . . . . . . . . . . . . . . 5

Data Explorer Source Profiler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Data Explorer FTM and Data Explorer XML . . . . . . . . . . . . . . . . . . . . . . 5

Data Explorer Repository Navigator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Data Explorer Repository . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Data Explorer DataDirect Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Chapter 2: Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Data Explorer Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Data Explorer Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Managing Temp Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Requirements for Column Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Requirements for Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Changing the Location of Temp Files . . . . . . . . . . . . . . . . . . . . . . . . . . 13

v

Managing Swap Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Managing Swap Files on UNIX Platforms . . . . . . . . . . . . . . . . . . . . . . . 15

Managing Swap Files on Windows Platforms . . . . . . . . . . . . . . . . . . . . . 15

Solaris Special Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Configuring Windows Server as an Application Server . . . . . . . . . . . . . . . . . 17

Chapter 3: Data Explorer Repository Administration . . . . . . . . . . . . .19

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Basic Repository Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Creating Synonyms and Aliases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Calculating Repository Disk Space Requirements . . . . . . . . . . . . . . . . . . . . . 22

Creating the Repository . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

ODBC and Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Setting Up a Data Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Sybase Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Oracle Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Informix Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

DB2 Universal Database Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Microsoft SQL Server Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Establishing Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

Setting the Environment Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

ODBC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

DB2 Universal Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Testing Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Chapter 4: Data Explorer Security Configuration . . . . . . . . . . . . . . . . 41

UNIX Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

User Privileges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

UMASK Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Windows Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Local Host Login . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Domain/User Name Login . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

User Impersonation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Client-Server Rights or Privileges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

Windows File Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

User and Group Authentication Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Project Authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

vi Table of Contents

Chapter 5: Data Explorer Listener Configuration . . . . . . . . . . . . . . . . 47

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Normal Startup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

Listener Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Listener Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

FireWall Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Security Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Services Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

Data Explorer Server Environment Variables . . . . . . . . . . . . . . . . . . . . . 52

Changing Port Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

License Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

License Server Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

License Tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

The License File (ELF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Finding a UMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Data Explorer Listener Web Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Listener Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

License Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Service Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

Inetd Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Responder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Starting and Stopping the Listener . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Start or Stop the Listener on Windows . . . . . . . . . . . . . . . . . . . . . . . . . 68

Start or Stop the Listener on UNIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Listener Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Chapter 6: Data Explorer Importer Configuration. . . . . . . . . . . . . . . . 71

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Customizing an Importer Configuration File . . . . . . . . . . . . . . . . . . . . . . . . 73

Configuration Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Customizing Flat-File Importers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

Flat-File Environment Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

DB2 Load Utility File Environment Variables . . . . . . . . . . . . . . . . . . . . 76

Customizing Informix Configuration Files . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Informix Environment Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Customizing Oracle Configuration Files . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Oracle Environment Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

Table of Contents vii

Customizing Sybase Configuration Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Sybase Environment Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Customizing Universal Database Configuration Files . . . . . . . . . . . . . . . . . . 81

Chapter 7: Data Explorer Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

Setting Project Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Operating System-based Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Listener-Based Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Deleting Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

Deleting Projects in the Data Explorer Client . . . . . . . . . . . . . . . . . . . . . 87

Deleting Projects from Data Explorer Server . . . . . . . . . . . . . . . . . . . . . 87

Compacting a Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Unlocking a Project on UNIX Platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Recovering a Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

Ensuring Project Integrity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Project Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Chapter 8: Data Explorer FTM/XML . . . . . . . . . . . . . . . . . . . . . . . . . . .93

System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Repository Connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

License Tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Installing the Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

User Administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

ODBC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

DTD Migration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Phase 1. Generate Mapping Report . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Phase 2. Generate DTD File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Phase 3. Reimport DTD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Phase 4. Recreate Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Chapter 9: Data Explorer Importer Administration . . . . . . . . . . . . . . 105

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

Data Explorer Source Profiler . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

viii Table of Contents

Data Explorer Import for VSAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

Appendix A: Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

Data Explorer Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

DLL Access Denied . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Disabled Autosync Flags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Cannot Reconnect to Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Data Explorer Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Server Connection Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

DBMS Connectivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Unable to Connect to DBMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Error While Exporting Project to Repository . . . . . . . . . . . . . . . . . . . . 114

License Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

License Server Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

No Licenses Available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

License Server Not Available . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

License Server Key File Invalid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

Temporary Key Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Resources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Kernel Exception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

Error Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Data Explorer Client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Data Explorer Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Operating System-Specific Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

Table of Contents ix

x Table of Contents

Preface

Welcome to Data Explorer Standard Edition, the Informatica software product suite that provides a complete and accurate picture of enterprise data through data profiling. Data Explorer is a complete set of data investigation, discovery, and mapping tools that allows you to quickly analyze multiple source systems and eases the process of scoping the size and complexity of a migration or data integration project.

xi

Informatica Resources

Informatica Customer PortalAs an Informatica customer, you can access the Informatica Customer Portal site at http://mysupport.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, Product Documentation, and access to the Informatica user community.

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 Product Documentation from http://mysupport.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://mysupport.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://mysupport.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,

xii : Preface

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://mysupport.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 SupportYou can contact a Customer Support Center by telephone or through the WebSupport Service. WebSupport requires a user name and password. You can request a user name and password at http://mysupport.informatica.com.

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

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

Toll Free Brazil: 0800 891 0202Mexico: 001 888 209 8853North America: +1 877 463 2435

Standard RateNorth America: +1 650 653 6332

Toll Free France: 00800 4632 4357Germany: 00800 4632 4357Israel: 00800 4632 4357Italy: 800 915 985Netherlands: 00800 4632 4357Portugal: 800 208 360Spain: 900 813 166Switzerland: 00800 4632 4357 or 0800 463 200United Kingdom: 00800 4632 4357 or 0800 023 4632

Standard RateBelgium: +32 15 281 702France: 0805 804632Germany: +49 1805 702 702Netherlands: +31 306 022 797

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

Standard RateIndia: +91 80 4112 5738

Informatica Resources xiii

xiv : Preface

C h a p t e r 1

Introduction

This chapter includes the following topics:

♦ Overview, 2

♦ Data Explorer Components, 3

1

Overview

Informatica Data Explorer includes:

♦ Data Explorer. Analyzes existing data sources to discover their content, structure, and quality, and produces a fully normalized data model. Data Explorer also allows users to design staging and target databases and to generate accurate source-to-target mapping and transformation specifications.

♦ Data Explorer FTM (Fixed Target Mapping). Maps profiled data sources to existing eCommerce applications, data warehouses, and packaged CRM and ERP systems.

♦ Data Explorer XML. An extension of Data Explorer FTM which maps profiled data sources to predefined XML standards using existing XML document type definitions (DTDs).

♦ Data Explorer Import for VSAM. For extracting OS/390 VSAM data and formatting it for data profiling in Data Explorer.

♦ Data Explorer Import for IMS. For extracting OS/390 IMS data and formatting it for data profiling in Data Explorer.

♦ Data Explorer Import for Relational. For connecting directly to an RDBMS for profiling the data in Data Explorer.

♦ Data Explorer Source Profiler. Used to prepare COBOL file definitions for extraction by Data Explorer Import for Files prior to importing the associated data into Data Explorer.

♦ Data Explorer Repository. Stores the information developed with Data Explorer data profiling and mapping and makes this information available for other Data Explorer products and for third-party data cleansing, movement and integration products.

♦ Repository Navigator. Provides a user-friendly interface for easy access to all information in the Data Explorer Repository.

Using a Windows client interface, Data Explorer users access server applications that run on UNIX or Windows platforms.

Users manage their own data, schemas, physical views, Data Explorer-generated reports, logs and DDL scripts, and to some extent projects. Data Explorer Project information can be exported to the Data Explorer Repository when the user wishes to make the Projects accessible to other users and applications.

Data Explorer Import users work with mainframe files and the Data Explorer Source Profiler Windows user interface. This interaction is described in detail in the Data Explorer Source Profiler User Guide.

The Data Explorer Repository is also the active, persistent data store for FTM/XML and Data Explorer Source Profiler. More information on these products is available in Data Explorer FTM/XML User Guide and Data Explorer Source Profiler User Guide respectively.

From an administrator's point of view, the Data Explorer includes various components, each of which may require discrete administration. This guide describes that administration.

2 Chapter 1: Introduction

Data Explorer Components

The Data Explorer components that a system administrator routinely interacts with include:

♦ Data Explorer Client

♦ Data Explorer Server

♦ Data Explorer Project

♦ Data Explorer Listener

♦ Data Explorer flat-file importers

♦ Data Explorer Import for VSAM

♦ Data Explorer Import for IMS

♦ Data Explorer Import for Relational Databases

♦ Data Explorer Source Profiler

♦ Data Explorer Repository

♦ Data Explorer DataDirect Technologies ODBC drivers

Data ExplorerData Explorer provides organizations with the ability to gain a thorough understanding of the content, structure, and quality of their corporate systems by analyzing actual source data. The Data Explorer discovery process reveals hidden data quality issues, inconsistencies, and incompatibilities between data sources and target applications. The result is an accurate description of the source data, which is stored in a central repository and is used to accelerate the design and implementation of new systems.

Data Explorer runs in a client-server configuration. The Data Explorer Client and Server can run on the same machine or the server can run on a different machine from the client. The remote client and server communicate through TCP/IP.

Copying or Moving Data ExplorerAfter installing Data Explorer, do not copy or move your installation using operating system commands. To move the Data Explorer application, always uninstall and then reinstall Data Explorer using the procedures described in this book.

The Data Explorer Client provides users access to most of the Data Explorer management tasks. The Data Explorer Client runs on Windows platforms. The client can run multiple instances of the server (up to the allocated number of licenses at the site).

Data Explorer ServerThe Data Explorer Server runs on or UNIX platforms. Administration tasks for the Data Explorer Server include:

♦ Changing the location to which temporary files are written during column profiling.

Data Explorer Components 3

♦ Managing swap space.

♦ Configuring the Data Explorer Server database connections, profiling options, and temporary directory locations.

For more information about server-related tasks, see “Resources” on page 7.

Data Explorer ListenerThe Data Explorer Listener monitors specific communication ports and starts the Data Explorer Server when a client attempts to make a connection to the server. Through a series of steps, the Listener starts the server and passes the server connection to the client. Data Explorer Listener is responsible for connecting a client to a Data Explorer Server and manages licenses as well.

Administration tasks for the Data Explorer Listener include:

♦ Changing port numbers

♦ Installing license key files

♦ Starting and stopping the Data Explorer Listener

For more information about these and other listening process-related tasks, see “Data Explorer Listener Configuration” on page 47.

Data Explorer ImportersUsers can import data into Data Explorer from a variety of sources. Data Explorer provides importers for:

♦ Flat files

♦ IBM DB2 LOAD utility files

Optionally, Informatica offers the following separately licensed importers:

♦ Data Explorer Import for VSAM. Captures data for profiling from VSAM databases and sequential files on IBM systems.

♦ Data Explorer Import for IMS. Captures data for profiling from IMS databases on IBM systems.

♦ Data Explorer Import for Relational Databases. Captures metadata and source data for profiling from database systems such as:

− DB2/UDB

− Informix

− Oracle

− Microsoft SQL Server (via ODBC)

− Sybase ASE

− Teradata


Each importer runs as a separate process. You can customize the environment for each importer using its associated configuration file. For information about changing importer environments, see “Data Explorer Importer Configuration” on page 71.

Data Explorer ProjectThe Data Explorer Project persistently stores the information for a particular Data Explorer project. Information enters the Project through Data Explorer Import, IROB format files, or through profiling and mapping tasks performed in Data Explorer. These processes are further described in the Data Explorer Orientation Guide.

Generally, users manage their own Projects via the Data Explorer Client. However, administration tasks for Projects include:

♦ Moving a Project

♦ Compacting a Project

♦ Unlocking a Project

♦ Recovering a Project

For more information about these and other Project-related tasks, see “Data Explorer Projects” on page 83.

Data Explorer Import for VSAM and IMSData Explorer Import for VSAM and Data Explorer Import for IMS are separately licensed products in the Data Explorer that prepare VSAM, IMS, and sequential data files from IBM systems for input to Data Explorer.

Data Explorer Import for VSAM and Data Explorer Import for IMS require little administration. For more information about these components, see “Data Explorer Importer Configuration” on page 71.

Data Explorer Import for Relational Databases is a separately licensed component of Data Explorer.

Data Explorer Source ProfilerData Explorer Source Profiler is a Windows application that uses COBOL data structures to help the user control the extraction process. Data Explorer Source Profiler uses the same concurrent process tokens that Data Explorer uses.

Data Explorer FTM and Data Explorer XMLData Explorer FTM (Fixed Target Mapping) maps profiled data sources to existing eCommerce applications, data warehouses, and packaged CRM and ERP systems. Data Explorer XML, an extension of Data Explorer FTM, which maps profiled data sources to predefined XML standards using existing XML document type definitions (DTDs). Using

Data Explorer Components 5

FTM/XML, you can map source metadata to target metadata and define the required data transformations.

Data Explorer Repository NavigatorRepository Navigator lets you view the contents of any Schema in the Repository. You can use Repository Navigator independently, or in conjunction with FTM/XML. Repository Navigator can be used simultaneously by any number of users.

Data Explorer RepositoryThe Data Explorer Repository is a relational database that stores the information contained in a Data Explorer Project. Users can export the contents of the Project to the Repository. Source Profiler, FTM/XML, and Repository Navigator use the Repository as their persistent store.

Because Repository information is stored in a relational database, it can be accessed using a variety of tools, such as report generators, interactive SQL, and so on. A detailed description of the Data Explorer Repository structure can be found in Data Explorer Repository Reference Guide.

Administration tasks for the Data Explorer Repository include:

♦ Creating a repository

♦ Setting up a data source

♦ Establishing connectivity

For more information about these and other repository-related tasks, see “Data Explorer Repository Administration” on page 19.

Data Explorer DataDirect DriversThe DataDirect ODBC Drivers for Data Explorer are included in the Data Explorer installation. For Windows, the DataDirect drivers you chose during installation (either version 5.2 or 5.3) are installed to the c:\winnt\system32 directory. The ODBC drivers can be used with the Data Explorer Server to access various databases. If you are planning to use a native driver, the DataDirect ODBC Drivers for Data Explorer are not required.

You can remove the DataDirect drivers using the Windows Add/Remove Programs feature.


C h a p t e r 2

Resources


♦ Overview, 8

♦ System Requirements, 9

♦ Managing Temp Space, 12

♦ Managing Swap Space, 15

♦ Configuring Windows Server as an Application Server, 17

7

Overview

Data Explorer products use various resources to operate successfully, including memory, swap space, and operating system components. If these resources are too limited or incorrectly configured, you may be unable to perform some Data Explorer functions. This chapter describes issues and administrative tasks related to resource issues and system requirements for the Data Explorer Server and Data Explorer Client.

For more information, see the Data Explorer Installation Guide.

8 Chapter 2: Resources

System Requirements

This section describes operating system and memory requirements for Data Explorer products.

Data Explorer ClientThe Data Explorer Client system requirements are:

♦ Windows

♦ 256 MB of RAM

♦ Pentium III or greater

♦ .NET Framework

♦ Screen resolution of 1024x768 or above (recommended)

The Data Explorer Client requires approximately 10 MB of disk space and, if the .NET framework is not installed, an additional 40 MB.

Data Explorer ServerThe Data Explorer Server system requirements are:

♦ Red Hat Enterprise Linux-based platform

♦ Sun SPARC-based Solaris platform

♦ IBM PowerPC-based AIX platform

♦ HP PA-RISC-based HP-UX platform

♦ Windows

A data profiling project requires a minimum of 512 MB of physical memory backed by 512-1024 MB of swap space. Large projects may require two to three times more memory and swap space. If you will be running other applications in addition to Data Explorer, additional memory and swap space may be necessary.

The Data Explorer Server requires approximately 130 MB of disk space.

Data Explorer Server Environment Variableside_server.ini sets the environment variables for the server execution environment. This file is modified by the installation process and is read by the Listener prior to running the Data Explorer Server.

System Requirements 9

Table 2-1 describes the environment variables within the ide_server.ini file that are configurable by the user:

Table 2-1. Data Explorer Server Environment Variables

Environment Variable Description

IDE_HOME Indicates the location of the Data Explorer Server installation.

IDE_ELMHOST Set this variable if the Data Explorer license server is running on a different subnet than the Data Explorer Server.

IDE_LS_WARN Specify the number of days before the expiration date when Data Explorer will begin issuing warning messages that the license will expire.

IDE_ELMPORT The port, if not 1416, where the license server is running.

IDE_LISTENER_PORT The port, if not 1416, where the local Listener is running, usually the same port as IDE_ELMPORT.

LD_LIBRARY_PATHLIBPATHSHLIB_PATH

Solaris and Linux library search path.AIX library search path.HP-UX library search path.

Path Windows library search path (in general, no need to set this variable).

TMP or TMPDIR Set this variable when you want to specify a location for the temporary files created by Data Explorer tasks such as column profiling.

IGNORE_FILE_LIST Defines a semi-colon list of directories and files that Data Explorer will ignore when processing a folder. This is useful to filter out slow network drives or to hide drives or specific folders.

IDE_UMASK Defines the umask settings for creating projects and files on UNIX. Set this if you want specific permissions turned on or off and specify the value in octal. For example, use “007” to turn off “other” permissions.

IDE_MAX_SAMPLE_SIZE Defines the maximum size of a sample data file that can be loaded into IDE. The default 64 MB, but the value can be set as low as 1 MB and as high as 1024 MB, assuming that there is enough memory allocated to the IDE server process. Setting this value over 512 MB may cause memory problems on your system.

IDE_COL_PROF_THREADS Defines the maximum number of threads to use for column profiling. The default is 1. Note that the IDE_COL_PROF_THREADS values in individual importer configuration files take precedence over the value in the ide_server.ini file.

IDE_COL_PROF_THREADS values cannot override the licensed number of processors. Setting values above the licensed number of processors has no effect.

IDE_COL_PROF_TMP_PATH Defines the temporary directories used by the importers in column profiling. Each path should be separated by a semicolon, e.g., temp1;temp2;temp3. The paths use round-robin cyclic scheduling to improve performance. Each importer is assigned a temporary directory and sets the appropriate TMP or TMPDIR path to it. For increased performance, each path should specify a directory on a separate physical disk drive.

If present, IDE_COL_PROF_TMP_PATH overrides the ide_server.ini file setting.


IDE_REPOSITORY_NAME_MAP Defines a name map file that maps repository column names to allowed values. Use this variable to implement the IDE repository on a non-supported RDBMS that specifies reserved column names. For example, if an RDBMS reserves the column name "Password", you can use a map file to specify an allowed value for the column name.

DEBUG_LOG_FILEDEBUG_LOG_LEVEL

Variables to help diagnose software problems.

Table 2-1. Data Explorer Server Environment Variables

Environment Variable Description

System Requirements 11

Managing Temp Space

On Windows platforms, the environment variable TMP determines where temporary files are created. If TMP is not defined, the root of the current drive is used.

On UNIX platforms, the environment variable TMPDIR determines where temporary files are created. If TMPDIR is not defined, /var/tmp or /tmp (in that order) is used. If you change this location by modifying TMP or TMPDIR, you should consider:

♦ Requirements for column profiling. For more information, see “Requirements for Column Profiling” on page 12.

♦ Requirement for Projects. For more information, see “Requirements for Project” on page 13.

♦ Choosing a disk that is not used for swapping to minimize disk contention and increase the performance of column profiling.

For information about changing locations, see “Changing the Location of Temp Files” on page 13.

When you change the location of temporary directories, make sure that Data Explorer users can write to the new location.

Note: If your Data Explorer installation permits multi-core processing, consider using the IDE_COL_PROF_TMP_PATH environment variable to set the paths to temporary directories used by the flat file importers in column profiling. For more information on IDE_COL_PROF_TMP_PATH, see Table 2-1 on page 10.

Requirements for Column ProfilingTemporary file space is needed primarily for column profiling from flat files. This section provides general guidelines for estimating the amount of temporary file space you will need to profile columns from flat files. (Data Explorer provides no way to specify where intermediate profiling results are put for the relational database importers, so consult your database administrator to configure temporary table space for relational databases.)

During the sort phases of column profiling, the intermediate results can be written to temporary files. In general, you need more than twice the space of the largest column that you are profiling. This formula will help you estimate the required space:

2 * (6 bytes + <MaxColumnSize>) * NumberOfRows

where 6 bytes is an overhead number for each value, MaxColumnSize is the largest size of all the columns in the file, and NumberOfRows is the number of rows in the file

For example, if you have a file whose largest column is 38 bytes wide containing 1,000,000 rows, the amount of temporary space needed to profile the file is 88,000,000 bytes (2 * 44 * 1,000,000).

If you do not know the MaxColumnSize, use the following general rules to determine how much temporary space you need.


♦ If the file contains 1 column (unusual case), the amount of temporary space needed will be approximately twice the size of the file.

♦ If the file contains 2 to 8 columns, the amount of temporary space needed will be approximately the size of the file.

♦ If the file contains more than 8 columns, the amount of temporary space needed will be approximately half the size of the file.

Requirements for ProjectProject space is required for storing Data Explorer objects.

♦ Metadata requires approximately 2 KB per Column.

♦ Use the following formula to estimate the amount of space sample data requires:

AverageLengthOfAValue * NumOfColumns * NumOfRows

♦ Use the following formula to estimate the amount of space profiling results requires:

(4 + AverageLengthOfAValue) * NumOfValues

Changing the Location of Temp FilesData Explorer allows you to choose a different location for temporary files.

To change the location of temporary files:

1. Edit the desired flat-file importer configuration files:

♦ idedb2file.ini. DB2 LOAD Utility importer configuration file.

♦ idefile.ini. Flat file importer configuration file.

On UNIX platforms, these files are located in:

/opt/Informatica/DataExplorer901/server901/config/fileimp

On Windows platforms, these files are located in:

Install_dir\Server\config\fileimp

2. Uncomment and modify the line containing the temporary directory variable to specify a temporary directory name. This line appears in the [Environment] section.

On UNIX platforms, this variable is TMPDIR and on Windows, the variable is TMP. On UNIX platforms, you also need to remove the pound sign (#). For example, to use the /data/temp as the temporary directory for the Data Explorer Flat File importer, change this line:

# TMPDIR=/tmp

to the following value:

TMPDIR=/data/temp

Note: Specify the full path name to the directory as environment variables will not be expanded.

Managing Temp Space 13

3. Make sure that all Data Explorer users have write permissions for the new location.


Managing Swap Space

The amount of virtual memory is controlled by the amount of available swap space (also called “paging space”) and is managed by the operating system.

Typically, the amount of swap space devoted to a computer is twice that of physical memory. For example, if your machine has 512 MB of physical memory, the swap space would be 1024 MB or 1 GB. This 2:1 ratio is approximate, and it may be acceptable to have much more space devoted to swap.

For performance reasons, you may want to spread your swap space over more than one disk. This allows the operating system to overlap writing and reading pages out to two or more disks.

Managing Swap Files on UNIX PlatformsTable 2-2 lists methods for managing swap space on each of the UNIX platforms supported by Data Explorer:

Managing Swap Files on Windows PlatformsWindows platforms allow you to define one swap file per disk. Each swap file has a minimum and maximum size.

To change the swap files on Windows platforms:

1. From the Windows Control Panel, double-click the System icon.

The System Properties dialog box appears.

2. Select the Performance tab.

Table 2-2. Commands to Manage Swap File Space

Platform Method Comments

Solaris swap command Use the “-l” flag to list swap files, the “-a” flag to add swap files, and the “-d” flag to delete swap files. For more information, see the Solaris documentation.To make these changes permanent, add lines to /etc/vfstab.Solaris also allows you to define a temporary file system that uses the swap file(s) as its storage mechanism. This sharing of swap and temp space requires special consideration. If you wish to take advantage of this feature, you may need to allocate additional swap space in addition to the expected virtual memory requirements to accommodate the large column profiling temporary files.

AIX smit and smitty tools Display the Paging Space dialog box by choosing Logical Volume Manager from the System Storage Management (Physical & Logical Storage) menu. For more information, see the AIX documentation.

HP-UX sam tool Use the Disk and File System dialog box. For more information, see the HP-UX documentation.

Managing Swap Space 15

On Windows 2000, select the Advanced tab and click Performance Options.

3. In the Virtual Memory section of the dialog, click Change.

The Virtual Memory dialog box appears.

4. Select the drive for which you wish to add a swap file.

5. Enter the Initial Size and Maximum Size for the swap file.

The initial size is the size of the swap file when Windows boots. The maximum size is the largest size the swap file can be.

In general, it is preferable to set the initial and maximum size to the same number as it is very inefficient for Windows to enlarge the swap file.

6. Repeat steps 4 and 5 for each disk drive to which you wish to add a swap file.

7. Click OK when done.

8. Reboot to have the new swap files take effect.

Solaris Special ConsiderationsBy default, the Solaris /tmp directory uses swap space. Because Data Explorer can also use /tmp, depending on the setting of the TMPDIR environment variable, you may run out of swap space when running memory-intensive processes or large column profiling tasks. To avoid this situation, change the location of the TMPDIR environment variable to one with adequate disk space. For more information about size estimates, see “Managing Temp Space” on page 12.


Configuring Windows Server as an Application Server

By default, a Windows server is configured to run as a file server instead of an application server. The default file server setting of Maximize Throughput for File Sharing gives the file cache priority in memory over applications. This allows the file cache to grow and consume physical memory, leaving insufficient space for applications to run and causing slowdowns as the applications thrash.

If you plan to run the Data Explorer Server on a Windows server, configure these servers to Maximize Throughput for Network Applications. This setting dramatically improves performance when reading files equal to or greater than the physical memory of the machine or when multiple applications are processing large amounts of I/O.

Configuring Windows Server as an Applicat ion Server 17

C h a p t e r 3

Data Explorer Repository Administration


♦ Overview, 20

♦ Basic Repository Requirements, 21

♦ Calculating Repository Disk Space Requirements, 22

♦ Creating the Repository, 24

♦ ODBC and Data Sources, 27

♦ Establishing Connectivity, 38

19

Overview

The Data Explorer Repository is a relational database that stores most of the information contained in a Data Explorer Project. The Repository is also the persistent store for the Data Explorer Source Profiler, Data Explorer FTM (Fixed Target Mapping), and Data Explorer XML. Because information is stored in a relational database, it can be accessed using a variety of tools, such as report generators and interactive SQL.

For more information about the Data Explorer Repository structure, see the Data Explorer Repository Reference Guide.

20 Chapter 3: Data Explorer Repository Administration

Basic Repository Requirements

Data Explorer uses Open Database Connectivity (ODBC) to connect to the database server where the Data Explorer Repository is installed. You must have a valid ODBC installation on the machine hosting the Data Explorer Server.

When choosing an ODBC vendor, consider the following two elements:

♦ The version of the ODBC API

♦ The conformance level of the ODBC driver

The Data Explorer applications require ODBC drivers to access the Data Explorer Repository.

ODBC drivers often use the database client access library to connect to the database server, so you may also need a valid database client installed for your ODBC driver to work. For example, if your target database server is Sybase, then an ODBC driver may use the Sybase CT library to connect to the Sybase server.

Creating Synonyms and AliasesOn some databases (such as UDB and Oracle), you may need to create synonyms or aliases for user access. To do this, query the system Project for tables that include “E01” and views that include “EV01.” You should be able to create synonyms or aliases and grant permissions as appropriate.

For example, to create synonyms in Oracle, you need to modify the select statement for your environment. Although this creates private synonyms, you can modify the statement for public synonyms and replace the xxxxxx and zzzzzzz with the user and owner's names. This also assumes that you have no other tables which have a name beginning with “E01,” so you might need to adjust for your environment. Then, you can take the output and run it to create your synonyms.

select 'create synonym xxxxxxxxx.'||object_name||' for '|| owner||'.'||object_name||';' from DBA_OBJECTS where owner = 'zzzzzzz' and object_type = 'TABLE' and object_name like 'E01%';

For more information about creating synonyms and aliases, see your the database documentation.

Basic Repository Requirements 21

Calculating Repository Disk Space Requirements

Use the following formula to calculate the disk space requirements of the repository:

RDBMS overhead

+ (RDBMS factor × [Data Explorer project size × Number of projects])

+ (Source Profiler schema size × Number of Source Profiler schemas)

+ (Data Explorer FTM schema size x number of schemas)

Contact your database administrator to determine the RDBMS overhead and RDBMS factor. To ensure that you have enough disk space for your repository, have your database administrator build tablespaces that you can expand later as per your requirements.

Table 3-1 describes the factors used in calculating the disk space requirements for the repository:

Consider the following example:

You install Data Explorer without Data Explorer FTM or Data Explorer Source Profiler and have only one Data Explorer project with a size of 35 MB. Your database administrator establishes that the RDBMS overhead is 13 MB and that a factor of 6 must be applied to any data loaded into the database. You can use the following formula to calculate the repository disk space requirements:

RDBMS factor 6 x Project size 35 MB 210 MB

Table 3-1. Disk Space Calculation

Term Meaning

Data Explorer project size Value calculated for the project’s metric.

Number of Data Explorer projects Maximum number of Data Explorer projects expected to be stored in the repository at any given time.

Number of Data Explorer FTM Schemas

Maximum number of schemas, originally created by the Data Explorer FTM product, stored in the repository at any given time.

Number of Source Profiler Schemas

Maximum number of schemas, originally created by the Source Profiler product, stored in the repository at any given time.

RDBMS factor Factor that adjusts the RDBMS storage overhead. The storage overhead is directly related to data that is loaded into the database. An RDBMS factor of 1 indicates that the RDBMS has no data-dependent overhead. A factor of 2 indicates that for every byte of data loaded into the repository, the RDBMS also consumes a byte of data-dependent storage.

RDBMS overhead Number of bytes consumed by the RDBMS installation after building the repository tables, and before loading any data into the database.

Data Explorer FTM Schema size Number of bytes expected to be stored in schemas created by the Data Explorer FTM/XML product.

Data Explorer Source Profiler Schema size

Number of bytes expected to be stored in schemas created by the Data Explorer Source Profiler product.


RDBMS overhead 13 MB

Total Estimated Disk Space for the Repository 223 MB

For information about managing temp space and swap space and other space considerations, see “Resources” on page 7.

Calculating Repository Disk Space Requirements 23

Creating the Repository

Data Explorer provides a set of SQL scripts to create the tables, indexes, and other repository objects. To execute these SQL scripts, you need a database user account with permissions to create tables, indexes, and foreign keys. Check with your database administrator to obtain a valid user account.

The SQL script files required for creating and maintaining the Repository are located in the following directories:

UNIX:

/opt/informatica/DataExplorer901/server901/sql

Windows:

Install_dir\Informatica Data Explorer\Repository\sql

Table 3-2 lists and describes the SQL script files:

To create a Repository, you must run these scripts in the order listed in the preceding table.

Note: If you use Sybase, you will receive two warnings about exceeding the maximum row size supported by the database. This is caused by one table with a few columns declared as VARCHAR (255). You can ignore these warnings. During the actual Repository operation this limit will not be exceeded.

If you use Oracle, you may need to edit the first INSERT statement in the setup.sql file to adjust the date format to match the date format of your environment.

For more information about possible issues when you run the repository scripts, see the Data Explorer Release Notes.

Table 3-2. SQL Script Files

SQL Script File Description

types.sql For Sybase and Microsoft SQL Server only. Creates new datatypes used by the Repository.

tables.sql Creates all the tables of the Repository.

indexes.sql Creates indexes on the Repository tables to optimize performance.

fkeys.sql Creates the foreign keys that enforce the referential integrity of the Repository.

setup.sql Initializes some of the tables with default values required by the Repository to work properly.

physprop.sql Initializes other Repository tables with default values required by Data Explorer export.

config.sql Configures the tables to accept custom values for tags. You can edit the script and use it to insert your own values, or run it to insert dummy records. You must execute this script when you create a Repository.

views.sql Optional. Creates the views on the base Repository tables. For more information about using Repository Views, see the Data Explorer Repository Reference Guide.


Setting Up the Repository for Custom Tag ValuesIn Data Explorer, some tag fields can contain custom values. Custom values match the needs of a particular project or site. To add custom values to the fields, edit the config\*.tag files on the Data Explorer Server. For the Repository to accept these custom field values, you must add the values to corresponding tables in the Repository. The easiest way to make these changes to the Repository is to replace the default values in the config.sql script with your custom values and then execute the script.

For example, the config.sql is shipped with placeholders values as shown in the following example:

insert into E01TransfType values (

29, '29'

)

;


30, '30'

)

;


31, '31'

So, to add the transformation types: “Huey”, “Louie”, and “Dewey”, change the lines as follows:


29, 'Huey'

)

;


30, 'Louie'

)

;


31, 'Dewey'

Creating the Repository 25

Repository Cleanup ScriptsUse droptables.sql to drops all tables, foreign keys, and indexes from the Repository database.

Warning: Executing this script removes the Repository and all of its stored information from the database. This operation is not reversible. The script does not prompt you to verify that you want to remove the Repository. Only use this script if you want to remove the Repository and all metadata from the database.

If you only want to remove the Repository Views, execute the dropviews.sql script.


ODBC and Data Sources

Before Data Explorer can connect to the database where the Repository has been created, you must install and configure ODBC on the computer that is running the Data Explorer Server.

Setting Up a Data SourceA data source is a set of parameters and information that ODBC uses to establish a connection with a database server. Multiple data sources can be associated with the same database server.

Although not absolutely necessary, it is strongly recommended that you create a data source for exclusive use with the Data Explorer Repository.

Windows SetupOn Windows platforms, data sources are described using the ODBC Data Source Administrator.

To set up an ODBC Data Source on Windows:

1. From the Windows Control Panel, select Administrative Tools-Data Sources (ODBC).

2. Click Add.

The Create New Data Source dialog box appears.

3. Select the driver you would like to use.

4. Follow the instructions.

You can modify your setting later by selecting the data source name and clicking Configure.

Note: Make sure you configure System DSN, as opposed to User DSN. Data Explorer recognizes only System Data Sources.

UNIX SetupThe information presented in this section assumes a DataDirect ODBC installation. If you are using a different product, refer to its documentation for specific details.

On UNIX platforms, data sources are described using a special file, odbc.ini. It is recommended that you use a centralized system information file that you control. Usually the ODBC installation supplies an example odbc.ini file in the ODBC installation directory. Here is an example:

[ODBC Data Sources]

Informix Wire Protocol=DataDirect Informix Wire Protocol

Oracle Wire Protocol=DataDirect Oracle Wire Protocol

SQLServer Wire Protocol=DataDirect SQL Server Wire Protocol

Sybase Wire Protocol=DataDirect Sybase Wire Protocol

ODBC and Data Sources 27

Teradata Wire Protocol=DataDirect Teradata Wire Protocol

[Informix Wire Protocol]

Driver=/opt/informatica/datadirect/lib/XOifcl17.so

Description=DataDirect Informix Wire Protocol

Database=db

HostName=informixhost

PortNumber=1500

ServerName=informixserver

EnableInsertCursors=0

GetDBListFromInformix=0

CursorBehavior=0

CancelDetectInterval=0

TrimBlankFromIndexName=1

ApplicationUsingThreads=1

The file contains a list of all data sources described in the file followed by an entry for each data source and its corresponding parameters. These parameters are specific to the database to which the data source connects.

To set up an ODBC Data Source on UNIX platforms:

1. Add a new entry under the [ODBC Data Sources] section of the file.

For example:

IDERelRep=Data Explorer Relational Repository using Oracle

2. Find a section in the odbc.ini file that corresponds to the database where the Repository is installed.

3. Copy the entire section to the end of the odbc.ini file.

4. Change the data source name of the newly created section to the same name that you used under the [ODBC Data Sources] section.

5. Configure the parameters for the data source.

These parameters are data-source dependent. The ODBC documentation describes all the parameters you can configure for each specific data source.

Note: While some ODBC drivers are self-contained and have all information inside the .odbc.ini file, most are not. For example, if you want to use an ODBC driver to access Oracle, you must install the Oracle SQL*NET software and set the appropriate environment variables. Verify such additional software configuration separately before using ODBC.


Sybase Data SourcesTable 3-3 describes the parameters you can configure for Sybase data sources:

Among the parameters you can configure for a Sybase data source, two are especially important:

Database=db

NetworkAddress=serverhost,4100

Oracle Data SourcesTable 3-4 describes the parameters you can define for Oracle data sources:

When configuring the data source, be sure to configure the server name. To connect to Oracle servers, use:

HostName=oracleserver

PortNumber=1521

SID=oraclesid

Table 3-3. Sybase Data Source Parameters

Parameter Description

NetworkAddress Required. Specify an IP address as follows: <servername or IP address>,<port number>For example, if your network supports named servers, you may specify an address such as:NetworkAddress=serverhost,4100You may also specify the IP address directly such as “199.226.224.34, 5000".

LoginID If required, type your login ID (case-sensitive).

Password If required, type your password for the system (case-sensitive).

Database Required. Type the name of the database you that want to access (case-sensitive).

Table 3-4. Oracle Data Source Parameters


HostName Type either the name or the IP address of the server to which you want to connect.

PortNumber Type the system port number for the Oracle listener.

SID Type the Oracle System Identifier that refers to the instance of Oracle running on the server. This option and the Service Name option are mutually exclusive. If the Service Name field is specified, do not specify this field.

ServiceName Type the database service name that specifies the database used for the connection. This option and the SID option are mutually exclusive. If the SID field is specified, do not specify this field.

Username If required, type your Oracle user name.

Password If required, type your Oracle password.


Informix Data SourcesIf you are using Informix, configure the following parameters in the odbc.ini file:

Database=db

HostName=informixhost

PortNumber=1500

ServerName=informixserver

Table 3-5 describes the parameters you can configure for Informix data sources:

Note: Data Explorer uses transactions during the export process; therefore your database must have logging enabled.

Also, when connecting to an Informix database, you must edit the informix.odbc file.

To edit the informix.odbc file:

1. Open the informix.odbc file in a text editor.

2. Locate the following line.

# IDE_INFORMIX_REP=1

3. Delete the “#” character preceding the line.

4. Save the edited file.

DB2 Universal Database Data SourcesThe IBM DB2 driver can be used to connect to DB2. DB2 requires that an ODBC driver is bound to a database, whether it is an IBM or OEM supplied driver. To bind a driver you need BIND authority on the database, so this is typically an administrator’s task.

If you use an IBM ODBC driver, the ODBC driver is installed (and bound) when defining a database in the DB2 control center. The control center performs the binds internally. (You can uncheck the option to install the ODBC driver connection but the default is to do this.)

With DataDirect drivers it is equally simple to do this. You set up the data source in the Windows ODBC Manager and, during the first connection (also a test connection), the plans are bound for the DataDirect driver in the database. This first connection may therefore take longer but it is a one time effort. The plan names are fixed within the DataDirect driver setup.

Table 3-5. Informix ODBC Parameters


Database Specifies the name of the database where the Repository has been installed.

HostName Specifies the name of the host where the Informix server runs.

PortNumber Specifies port number where the Informix server runs.

ServerName Name of the Informix server, typically in the form server_on_net.


This means that this has to be done really just once, and then all users using the same driver have instant access to the database. Be aware that you may receive error messages such as “BIND FAILED” if you are not authorized to bind the plans. Check with your DBA for assistance.

You can also refer to the DataDirect documentation for more information.

If you are using Solaris or AIX, you will need IBM DB2 Client Application Enabler, as well as the IBM DB2 Software Development Kit installed. The DB2 server has to be installed as the Server Version, not the Local Version.

You also need to configure the following parameters in your odbc.ini file:

Database=database_name

HostName=host

ServerName=server

Table 3-6 describes the parameters you can edit for IBM DB2 data sources:

Table 3-7 describes the DataDirect parameters you can edit for IBM DB2 data sources:

Table 3-6. IBM DB2 ODBC Parameters


Database Specifies the name of the database where the Repository is installed. If you are running DB2 on Windows or UNIX, type the name of the remote database that you want to access. You must have cataloged any database that you wish to access from the client.

HostName Specifies the name of the host where the Repository is installed.

ServerName Specifies the name of the server where the Repository is installed.

Table 3-7. IBM DB2 Data Direct Parameters


IpAddress Type the IP (Internet Protocol) address of the machine where the catalog tables are stored. Specify the address using the machine's numeric address (for example, 123.456.78.90) or specify its host name. If you enter a host name, the driver must find this name (with the correct address assignment) in the HOSTS file on the workstation or in a DNS server.

TcpPort Type the port number that is assigned to the DB2 server on the machine where the catalog tables are stored. Specify either this port's numeric address or its service name (5179 is the default port address). If you specify a service name, the driver must find this name (with the correct port assignment) in the SERVICES file on the workstation.

Location Type the DB2 location name for some versions of DB2. Use the name defined during the local DB2 installation.

Collection Type the name that identifies a logical group of database objects for some versions of DB2. For some versions, it is also the current Schema. These objects include the DataDirect Connect for ODBC DB2 driver packages.

LogonID If required, type your user name (logon ID).

Password If required, type your password.


Microsoft SQL Server Data SourcesUse the following steps as a guideline to configure ODBC in Windows and UNIX. For more information about specific connectivity issues, see the Microsoft SQL Server documentation.

To set up ODBC data source on Microsoft SQL Server in Windows:

1. Install the Microsoft SQL Server client and configure native connectivity.

2. From the Windows Control Panel, select ODBC.

The ODBC Data Source Administrator utility appears.

3. Click Add.

The Create New Data Source dialog box appears.

4. Select SQL Server from the scrolling list.

Note: Do not use the DataDirect 32-bit closed ODBC driver for Microsoft SQL Server provided by Informatica.

5. Click Finish.

SecurityMechanism Select the security method the driver uses to authenticate the user to the DB2 server when a connection is established.0 - No Encryption (the default), the driver sends the password in clear text to the DB2 server for authentication.1 - Encrypt Password, the driver sends an encrypted password to the DB2 server for authentication. 2 - Encrypted UID and Password, the driver sends an encrypted user ID and password to the DB2 server for authentication.If the specified authentication method is not supported by the DB2 server, the connection fails and the driver generates an error.

Database Type the database name of the DB2 database.

Table 3-7. IBM DB2 Data Direct Parameters



The Create a New Data Source in SQL Server dialog box appears.

6. Enter the name and description of the data source.

7. Select the SQL Server from the Server list.

8. Click Next.

Options appear for verifying the authenticity of the login ID.

9. Select With SQL Server Authentication Using a Login ID and Password Entered by the User.

This login ID and password correspond to a valid account on the Microsoft SQL Server you want to use.

10. Click Next.


Options appear for selecting the default database and setting other parameters.

11. Change the default database to the database where you have installed the repository.

To ensure consistent data in Microsoft SQL Server repositories, clear the Create temporary stored procedures for prepared SQL statements option.

To ensure that you can perform column profiling using SQL Server, select the Use ANSI quoted identifiers and Use ANSI nulls, paddings, and warnings options.

12. Click Next.

Retain the default settings in the dialog box that appears.

13. Click Finish to complete the ODBC setup.

To set up ODBC data source on Microsoft SQL Server in UNIX:

1. To configure connectivity for the Integration Service process, log in to the machine as a user who can start a service process.

2. Set the ODBCHOME and PATH environment variables.

♦ ODBCHOME. Set to the DataDirect ODBC installation directory. For example, if the install directory is /opt/ODBC5.1.

Using a Bourne shell:

$ ODBCHOME=/opt/ODBC5.1; export ODBCHOME

Using a C shell:

$ setenv ODBCHOME /opt/ODBC5.1

♦ PATH. To run the ODBC command line programs, like ivtestlib, set the variable to include the odbc bin directory.


$ PATH=${PATH}:$ODBCHOME/bin; export PATH


Using a C shell:

$ setenv PATH ${PATH}:$ODBCHOME/bin

Run the ivtestlib utility to verify that the UNIX ODBC manager can load the driver files.

3. Set the shared library environment variable.

The ODBC software contains a number of shared library components that the Integration Service and Repository Service processes load dynamically. To locate the shared libraries during run time, set the shared library environment variable.

The shared library path must also include the Data Explorer installation directory (server_dir).

Set the shared library environment variable based on the operating system. The following table describes the shared library variables for each operating system:

For example, use the following syntax for Solaris:

♦ Using a Bourne shell:$ LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:$HOME/server_dir:$ODBCHOME/lib; export LD_LIBRARY_PATH

♦ Using a C shell:$ setenv LD_LIBRARY_PATH $HOME/server_dir:$ODBCHOME:${LD_LIBRARY_PATH}

For HP-UX:

♦ Using a Bourne shell:$ SHLIB_PATH=${SHLIB_PATH}:$HOME/server_dir:$ODBCHOME/lib; export SHLIB_PATH

♦ Using a C shell:$ setenv SHLIB_PATH ${SHLIB_PATH}:$HOME/server_dir:$ODBCHOME/lib

For AIX:

♦ Using a Bourne shell:$ LIBPATH=${LIBPATH}:$HOME/server_dir:$ODBCHOME/lib; export LIBPATH

♦ Using a C shell:$ setenv LIBPATH ${LIBPATH}:$HOME/server_dir:$ODBCHOME/lib

4. Edit the existing odbc.ini file or copy the odbc.ini file to the home directory and edit it. This file exists in $ODBCHOME directory.

$ cp $ODBCHOME/odbc.ini $HOME/.odbc.ini

Operating System Variable

Linux LD_LIBRARY_PATH

Solaris LD_LIBRARY_PATH

AIX LIBPATH

HP-UX SHLIB_PATH


Add an entry for the ODBC data source under the section [ODBC Data Sources] and configure the data source.

For example:

MY_MSSQLSERVER_ODBC_SOURCE=<Driver name or Data source description>

[MY_SQLSERVER_ODBC_SOURCE]

Driver=<path to ODBC drivers>

Description=DataDirect 5.2 SQL Server Wire Protocol

Database=<SQLServer_database_name>

LogonID=<username>

Password=<password>

Address=<TCP/IP address>,<port number>

QuoteId=Yes

AnsiNPW=Yes

ApplicationsUsingThreads=1

This file might already exist if you have configured one or more ODBC data sources.

5. Verify that the entries QuoteID and AnsiNPW in the odbc.ini file are set to Yes.

6. Verify that the last entry in the odbc.ini is InstallDir and set it to the odbc installation directory. For example:

InstallDir=/usr/odbc

7. If you use the odbc.ini file in the home directory, set the ODBCINI environment variable.


$ ODBCINI=/$HOME/.odbc.ini; export ODBCINI

Using a C shell:

$ setenv ODBCINI $HOME/.odbc.ini

8. Edit the .cshrc or .profile to include the complete set of shell commands. Save the file and either log out and log in again, or issue the source command.


$ source .profile

Using a C shell:

$ source .cshrc

9. Use the ivtestlib utility to verify that the UNIX ODBC manager can load the driver file you specified for the data source in the odbc.ini file.

For example, if you have the driver entry:

Driver = /opt/odbc/lib/DWxxxx.so

issue the following command:


ivtestlib /opt/odbc/lib/DWxxxx.so

10. Install and configure any underlying client access software needed by the ODBC driver.

Table 3-8 describes the parameters in the odbc.ini file for Microsoft SQL Server data sources on UNIX:

Table 3-8. Microsoft SQL Server ODBC Parameters on UNIX


Address The network address of the server running Microsoft SQL Server. This is required and can be an IP address in the following format: IP address, port_numberFor example, you might enter 199.226.224.34, 5000. If your network supports named servers, you can specify an address as: servername, port_number. For example, you might enter SSserver, 5000. To specify a named instance of Microsoft SQL Server, use the format: server_name\instance_name.If only a server name is specified with no instance name, the driver connects to the server and uses the default named instance on the server.

Database The name of the database to which you want to connect by default. If you do not specify a value, the default database defined by Microsoft SQL Server is used.


Establishing Connectivity

Besides configuring the data sources, ODBC for UNIX also requires that you set some environmental variables before you can establish a connection with the server. These variables are contained in the ide_server.ini file that starts Data Explorer.

This section describes how to make the changes to the file. This section also includes the environment variables required by each data source, as well as shows you how to test your data source.

This section contains references to the library path variable. This is the operating system-specific variable that defines the search path for shared libraries.

♦ For Solaris and Linux, this is LD_LIBRARY_PATH

♦ For HP-UX, this is SHLIB_PATH

♦ For AIX, this is LIBPATH

When changing the library path you should always preserve what was there before. Here is an example using LD_LIBRARY_PATH and Bourne shell:

LD_LIBRARY_PATH=$(LD_LIBRARY_PATH):other directories

Setting the Environment VariablesData Explorer includes a set of files that contain the environment variables that you must set to connect to your database of choice. There is one file per supported database, plus one file for the ODBC variables, as shown below.

♦ Database-specific file. Environment variables defined for a particular database are located in:

/Install_dir/server901/DatabaseType/Database.odbc

♦ ODBC file. Definitions required by ODBC for any database are located in:

/Install_dir/server901/odbc/odbc.odbc

If you look at any of these files, you can see that several lines are commented out with a “#” character. These lines contain environment variables that must be defined for that particular database or for ODBC in the odbc.odbc case.

Edit the odbc.odbc file, as well as the files that correspond to the databases to which you plan to connect, to make sure they define variables that match your environment. As part of this process, you must:

1. Uncomment the lines that contain the variables that must be defined.

2. Make sure that you define values that match your environment.

Note: The values that are predefined in these files are only generic examples of what these values could be.


ODBCDataDirect Technologies ODBC drivers require that two variables be set.

♦ The library path variable must point to the lib directory that is in the ODBC installation directory. For example, if ODBC is installed in /opt/informatica/datadirect:

LD_LIBRARY_PATH=$(LD_LIBRARY_PATH):/opt/informatica/datadirect/lib

♦ The ODBCINI variable must point to the odbc.ini file you plan to use. For example, if ODBC is installed in /opt/odbc:

ODBCINI=/opt/informatica/datadirect/odbc.ini

For more information, please see “Setting Up a Data Source” on page 27.

DB2 Universal DatabaseDB2 Universal Database requires that three environment variables be set:

♦ The DB2HOME variable must point to the DB2 installation directory

♦ The DB2INSTANCE variable must include the name of the DB2 instance

♦ The library path variable must include the $(DB2HOME)/sqllib/lib directory

Testing ConnectivityIt is always a good idea to verify whether you can establish connection with a given data source before trying to access it through Data Explorer.

The file, ivtestlib, is in the /bin directory and you can use it to make sure that you have the environment variables correctly set and that all shared libraries necessary for the driver operation can be properly loaded.

To test an ODBC driver:

1. Set the environment variables described in the previous sections.

2. From the ODBC installation directory, execute the command:

bin/ivtestlib driver_name

where driver_name is the name of the driver (not the name of the data source) as it appears in your odbc.ini file. For example, if your data source name is Sybase and your odbc.ini file contains:

[Sybase]

Driver=/opt/odbc/lib/ivsyb13.so

Description=Sybase

Database=odbc

ServerName=SYBASE

WorkstationID=id

LogonID=odbc01

Establishing Connectivity 39

Password=odbc01

OptimizePrepare=2

SelectMethod=1

then execute the command:

bin/ivtestlib ivsyb13.so

3. To test the utility to make sure the ODBC shared library can be properly loaded, execute the command:

bin/ivtestlib libodbc.so

This example is for Solaris and AIX. On HP-UX platforms, the name of the ODBC library is libodbc.sl.


C h a p t e r 4

Data Explorer Security Configuration


♦ UNIX Security, 42

♦ Windows Security, 43

♦ User and Group Authentication Models, 45

41

UNIX Security

This section describes security and user privilege issues related to running Data Explorer on UNIX platforms.

User PrivilegesYou can install Data Explorer as a non-root user or as a root user. If you install as a non-root user, the Data Explorer Listener Web Interface provides authentication permissions for Data Explorer projects. If you install as a root user, the user permissions on the Unix server are used for project authentication permissions.

Installing as a Non-Root UserNon-root users can install and manage Data Explorer. For non-root installations, the Data Explorer Listener Web Interface provides the project authentication permissions for the repository. For more information, see the authentication administration settings in “Administration” on page 58.

To ensure access to the Data Explorer listener after a machine reboot, follow the steps in “Creating Listener Startup and Shutdown Scripts” in the Data Explorer System Administration Guide.

Installing as a Root UserIf you choose to install as a root user, any user who wants to use Data Explorer on a UNIX server must have a user name that can be validated by the UNIX server.

UMASK IssuesBy default, the application permissions that Data Explorer uses are open, that is, read, write, and execute for user, group, and other. Generally, UNIX allows an individual’s file permission defaults (as set in the user’s .profile or .cshrc file) to override the application’s default permissions. However, because Data Explorer runs through the Listener, it does not read a user’s startup file. This means that Data Explorer cannot pick up any individual UMASK settings that the user designates. However, Data Explorer does pick up the root startup UMASK, with the defaults to read, write, and execute for user and read and execute for group and other.

On the IBM AIX platform, Data Explorer picks up the UMASK settings if they appear in /etc/security/user.

To override the root default UMASK settings, edit the ide_server.ini file, which is in the /Install_Directory/server/bin directory, and remove the pound sign (#) to uncomment the following line:

# UMASK=000

42 Chapter 4: Data Explorer Security Configuration

Windows Security

This section describes security and user privilege issues related to running Data Explorer on Windows platforms.

Local Host LoginWhen logging into a host, a Data Explorer user has three options for the host name in the Login dialog box:

♦ IP address

♦ Node (machine) name

♦ The string “localhost” (if the client and server are on the same machine)

Domain/User Name LoginIf your site has multiple domains, or if machine names and user names can be the same string, you should always log in to Data Explorer with your <domain name>\ <user name> as the user name.

User ImpersonationWhen connecting from a Data Explorer Client to a Data Explorer Server, a user must supply his or her user name and password. After the server validates the user name and password, the server changes from a system process to a user process and “impersonates” the user, that is, the server has the same rights and privileges as the user.

Client-Server Rights or PrivilegesIf you want users to run the Data Explorer Client against a remote Windows server, you must grant logon privileges to any Data Explorer user who connects to the remote Windows server.

Data Explorer users must also have write access to the Windows server so the server can create Projects and session and log files. This is because the server becomes a user process.

To accomplish this, the system administrator must grant Log on Locally privileges to a user or a group of users.

To grant the Log on Locally privilege on Windows:

1. From the Windows Control Panel, select Administrative Tools.

The Administrative Tools directory appears.

2. Double-click Local Security Policy.

The Local Security Settings window appears.

3. Expand the Local Policies directory.

Windows Security 43

4. Double-click User Rights Assignment.

The Policy window is populated with user right policies.

5. Double-click Log on Locally.

The Local Security Policy Setting dialog box appears.

6. Click Add.

The Select Users and Groups dialog box appears.

7. Click Show Users.

8. Select from the Name field the users or groups to whom you wish to grant access privileges.

9. Click Add.

10. Click OK.

11. In the User Rights Policy dialog box, click OK.

Rebooting the Windows server is not necessary.

Windows File SystemsIn a shared File Allocation Table (FAT) system, Data Explorer does not need to observe security when writing a Project. This is because a FAT file system does not implement the file access control lists that NT File System (NTFS) does. In NTFS, Data Explorer observes all security settings of NTFS when writing a Project.


User and Group Authentication Models

When working with users and groups, take note of the following authentication conditions:

♦ A root install uses OS authentication and file system enforcement. To open a project in a root installation, you must have read and write privileges to the project. A root user or Administrator can access any readable/writable project.

♦ A non-root install uses Data Explorer listener authentication, which mimics the UNIX authentication model. UNIX authentication enforces the following rules for user and group names:

− A name can have a maximum of 80 characters.

− Names are case-insensitive.

− UNIX removes leading and trailing spaces are removed from user and group names. It preserves internal spaces.

− Names may not contain these special characters:

, + " \ < > ; / * % ? @

− User names must be unique within the list of users.

Administrators can open and manipulate any project created by a user. For root installs, administrators can perform administrative tasks in the Data Explorer Listener, such as terminating the Data Explorer Server and removing licenses.

You can optionally define groups for your users. The following conditions apply to groups:

♦ Group names are semi-colon separated.

♦ Groups have a different name space to Users: User “JohnDoe” and Group “JohnDoe” are not the same object.

♦ New Data Explorer projects default to the first-created group. Use the Data Explorer user interface to change to any group to which you belong or to disable the default setting.

Project Authentication The following conditions apply to project authentication:

♦ The project must be readable and writable by the user. If not, authentication fails.

♦ An administrators can open all projects and change the owner and group for any project. An Administrator can add a project to any defined group in the Listener or can remove the project from group ownership.

♦ A project owner can open each of their projects. A project owner can add a project to any defined group in their list or remove the project from group ownership.

Non-owners can open a project that has been added to one of their groups.

User and Group Authentication Models 45

C h a p t e r 5

Data Explorer Listener Configuration


♦ Overview, 48

♦ Data Explorer Listener Web Interface, 57

♦ Listener Configuration, 50

♦ License Server, 53

♦ Starting and Stopping the Listener, 68

♦ Listener Components, 69

47

Overview

When a user logs in to the Data Explorer from a client, the client connects to the Data Explorer Server through the Data Explorer Listener.

The Data Explorer Listener is a multi-threaded system service that runs on Windows or UNIX and is responsible for launching the Data Explorer Server and issuing license keys that control access to Data Explorer. The Listener runs in the background, listening for connections on a well-known port. When a request comes in, the Listener services the request by either starting a Data Explorer Server session or responding to license key requests.

Normal StartupThe Listener normally starts automatically when the computer boots and continues to run as long as the computer runs. When it first starts up, the Listener reads a configuration file that controls its settings and may also read an optional license file. It monitors these files for changes and applies them immediately. Therefore, the Listener need not be restarted as part of its normal operation.

Figure 5-1 shows how the Listener starts the Data Explorer Server and how a web browser can monitor its activities:

Figure 5-1. Data Explorer Listener Architecture

Data Explorer Client

Browser

listener.ini

ide_server.ini4

7

2

56

3

1

Data Explorer Server

1416Data Explorer Listener

Client Machine Server Machine

Web Browser

48 Chapter 5: Data Explorer Listener Configuration

1. The Listener starts up and reads the listener.ini file, which contains a list of Data Explorer Servers and their versions.

2. The Data Explorer Client connects to the Listener on its well known port and requests that it starts a Data Explorer Server that has a compatible version.

3. The Listener reads ide_server.ini to set the environment before launching the Data Explorer Server.

4. The Data Explorer Client then connects to the Data Explorer Server and starts issuing it commands and receiving data.

5. Periodically, the Data Explorer Server sends the Listener status information to be displayed by the web interface.

6. Optional. If allowed by the Listener, the user may terminate the Data Explorer Server or cancel one of its jobs. The Listener will then connect to the Data Explorer port and issue a terminate command, which will cause the Data Explorer Server to stop immediately.

7. Optional. A web browser may connect to the Listener on its well known port number and review information posted by the Data Explorer Server or issue it commands, such as terminate or Cancel job.

Overview 49

Listener Configuration

The installation script performs most of the configuration needed to prepare the Listener for processing messages and starting the Data Explorer Server. This section details settings you can modify in the Listener configuration file when necessary.

When the Listener first starts up, it reads the listener.ini file located in the Data Explorer Listener directory. On UNIX (assuming you installed in the default directory):

/opt/Informatica/DataExplorer[VersionNumber]

On Windows:

C:\Informatica\DataExplorer[VersionNumber]

Once read, the Listener monitors this file for changes and applies the changes immediately.

The listner.ini file is divided into four major sections described below along with the parameters in each section.

Listener SectionThis section defines general operating parameters for the Listener. In general, these parameters do not need to be changed:

♦ ListenerPort. Set to the TCP/IP port number on which the Listener listens for commands and web requests. Change this only if it conflicts with another application on the server computer.

If you change the ListenerPort, ensure that users logging in to the Data Explorer Client change their port address to the correct setting.

♦ LogFileSize. The Listener creates a log file of errors and activities in the file, “listener/log/listener.log”. This parameter sets the maximum size of the file before it is copied to listener.log.bak and cleared. If listener.log.bak exists, it is overwritten when the listener.log rolls over.

♦ LicenseManager. Sets a flag to enable or disable license management. In general, you can leave this option enabled. The default license file, license.elf, that ships with the Listener contains no tokens so it effectively turns off license management.

♦ LicenseFile. This file contains the license tokens that the Listener manages. Since the default license file contains no tokens, if you get a new license file from Informatica Global Customer Support, ensure that the path points to the license file. The Listener monitors this file for changes, so if you overwrite the current license file with a new one, the Listener immediately applies the changes.

♦ ZombieCheck. This defines the number of seconds that the Listener waits for the Data Explorer Server to update itself with changes to the Listener before the server is removed from the active session list.


FireWall SectionThis section defines the parameters for allocating ports to the Data Explorer Server. By default, the Data Explorer Listener allows the operating system to choose a random port to assign the Data Explorer Server when it first starts up.

Some installations require the Data Explorer Client to sit behind a firewall. In this case, the Listener can be configured to allocate ports from a specific range. Generally, the number of ports that need to be opened up is the number of Data Explorer tokens plus the number of Data Explorer Batch tokens:

♦ FirstServicePort. The first port number in the inclusive range of numbers to open up. Most operating systems reserve the low port numbers, so set this number at a value greater than 1000.

♦ LastServicePort. The last port number in the inclusive range of numbers to open up.

Note: You must allocate two ports for every license token. For example, if you have 10 license tokens to run 10 concurrent server sessions, you must allocate 20 ports.

Security SectionThis section defines the web interface security policy. The web interface has two areas, the informational user area and an administrative area. The Listener controls these areas by allowing full access, password access, or no access. The following table describes the logical setting for these security parameters.

If there is no password file detected when the Listener first starts up, the Listener creates a new one. The file is encrypted and it contains a single administrator entry without a password. If the administrator password is forgotten, you must delete the password file and re-create the password entries using the web interface.

♦ PasswordFile. Location of the Listener password file. It must be created by a user with root or Windows Administrator privileges.

♦ User. Specifies the user area security policy as Allow, Password, or Deny.

♦ Admin. Specifies the administrative area security policy as Allow, Password, or Deny.

Services SectionThis section defines the services that the Listener can start. The variables in this section refer to named sections that define the details of the service.

User Area

Allow Password Deny

Admin Area

Allow Not Recommended Questionable Invalid

Password Recommended OK Invalid

Deny OK OK OK

Listener Configuration 51

Data Explorer Server Environment VariablesWhen the Data Explorer Listener starts a Data Explorer Server, the Listener reads the environment variables specified for the server in the ide_server.ini file and sets these variables in the environment. To change environment variables, edit this file.

Changing Port Numbers By default, the Data Explorer Listener uses the port address of 1416. If necessary you can change the port number in the listener.ini file, for example, to resolve a conflict with existing applications. However, if you change the default port number, be sure to also change the port number on the Data Explorer Client and on all applications that use the Listener as a license server. You do this by updating the IDE_ELMPORT environment variable.

Note: You must have administrative privileges on the machine to set the port numbers on Windows platforms.


License Server

The Listener contains an embedded license server that uses a feature based licensing scheme. Each feature contains a pool of tokens that represent the concurrent number of applications that can access the feature. When the application requires a token, it sends a message to the license server requesting a token for the feature. If there is a free token, the token is issued to the application. Otherwise, the request is denied and the application refuses to perform the task.

The following features are currently licensed by the Data Explorer family of products:

License Server ArchitectureThe license server is based on the client/server model with:

♦ The license server running on one machine

♦ The license clients running on one or more machines

The license clients and the Data Explorer Listener license server may be on the same machine or they may be on different machines, as in a network environment. Each machine on the network has a Unique Machine Identifier (UMI), which can be used to lock the license server or the applications to the machine.

License TokensA token contains information about when and where it can be allocated. This information is:

♦ Begin Date. The date, which is one second after midnight the day before, when the token can first be allocated.

♦ End Date. The date, one second before midnight the day of, when the token can be last allocated. A special flag, “Perpetual,” means there is no end date.

♦ Restrictions. The access control list for the token. The token may be restricted to certain clients based on the client UMI or application.

Number Name Feature Licensed

1 BASE Shared by Data Explorer Server and Data Explorer Source Profiler

2 RDBMS Data Explorer Import for Relational Databases

3 FTM Data Explorer FTM (Fixed Target Mapping)

4 XML Data Explorer XML

5 DataDirect Data Explorer DataDirect Technologies ODBC Drivers

6 Batch Data Explorer Batch Server

License Server 53

The License File (ELF)The license server requires a license file (license.elf ), sometimes called the “ELF” file, to indicate which features are licensed and for how long. The license file is generated by Informatica Global Customer Support using the following information:

♦ Unique Machine Identifier (UMI) of the Data Explorer Listener machine

♦ Start date

♦ End date

♦ Number of tokens

♦ List of licensed machines on which the Data Explorer Server can run

♦ List of components that you have been licensed to run

Because the UMI and start and expiration dates are included in the license file, the license server can operate only on the designated machine and issues tokens only during the designated time period. The token count and allocation may be changed by installing a new license file.

Obtain and Install a License FilePerform the following steps to obtain and install a license file:

1. Find the UMI of the machines on which you intend to run the license server and the Data Explorer Server. For more information on finding a machine UMI, see “Finding a UMI” on page 55.

2. Provide the UMIs and request a license file from Informatica Global Customer Support. You will receive a license file by email.

3. Save the license file to disk.

4. Copy the license file to the Listener machine in the “listener” directory of the Data Explorer installation.

The Listener will automatically read this file and update the license server’s internal token database with the new features.

Different SubnetsTo run the Data Explorer Server and the Listener on different subnets requires special set up because the Data Explorer Server initially does a broadcast looking for a license server. Broadcast messages do not cross subnet boundaries. To remedy this, you must set the IDE_ELMHOST environment variable on the ide_server.ini file.

To set IDE_ELMHOST:

1. Edit the ide_server.ini file in the Data Explorer Server bin directory.

2. Look for the following line.

#IDE_ELMHOST=@localhost


Note: The @ symbol is a required syntactical element.

3. Uncomment out the variable by deleting the pound sign (#).

4. Replace “localhost” with the name or IP address of the machine running the license server.

5. Save the edited script.

Setting IDE_ELMHOST for Other Data Explorer ComponentsWhen running Repository Navigator, Source Profiler, and FTM/XML, you must also set the IDE_ELMHOST environment variable on the client system, so these programs can communicate with a license server.

Repository Navigator, Source Profiler, and FTM/XML do not have a configuration file and therefore look to your system environment variable for an instance of that variable/value. If this value is not present, these programs cannot locate the license server.

To create the IDE_ELMHOST variable:

1. On the local machine, right click on My Computer and select Properties.

2. In the Properties dialog box, select Advanced > Environment Variables > System Variables.

3. Click the New button.

4. Type the Variable Name:

IDE_ELMHOST

5. Type the Variable Value:

@localhost

6. If the license server port number has changed from the default (1416), type the Variable Name:

IDE_ELMPORT

7. Set the value to 1416.

Finding a UMIThe UMI for each machine is a series of digits and/or letters in the correct format needed by Informatica Global Customer Support to generate a license file.

Note: The setup.csh script provides an option for printing the UMI.

The preferred method of finding a UMI is to run the Data Explorer setup.csh script for UNIX or running the Data Explorer UMI Generator for Windows. However, you can find your UMI using the command for your platform, as shown in the following table. Use the command exactly as shown; do not add flags to or remove flags from the commands.

License Server 55

Table 5-1 describes the commands for finding a UMI:

The unique machine identifier is a series of digits and/or letters in the correct format needed by Informatica Global Customer Support to generate a license file.

To use the Data Explorer UMI Generator in Windows:

From the Data Explorer program group, click Data Explorer UMI Generator. A dialog appears showing the UMI needed to generate a license file for your machine.

Click OK.

Table 5-1. Commands to Find the UMI (Unique Machine Identifier)

Platform Command

HP-UX /usr/bin/uname -i

AIX /usr/bin/uname -l

Solaris /usr/bin/hostid

Linux <IDEInstallDir>/listener/bin/./idelistener -umi

Windows Use the Data Explorer UMI Generator.


Data Explorer Listener Web Interface

The Data Explorer Listener web interface allows you to access the Listener from a web browser to change or view Listener settings and to manage Data Explorer licenses and sessions.

Using the web interface, you can generate reports, administer the Data Explorer Listener, and check the status of the Listener services. Depending on the security settings and access level, users can view status or make changes in two different areas:

♦ The informational area that displays the status of the Listener, licenses, and Data Explorer sessions. This area is read-only.

♦ The administrative area where the user can modify the state of processes that the Listener manages.

To open the Data Explorer Listener web interface:

Open a browser to the following URL:

http://machine:listenerPort

For example, http://192.168.1.123:1416

-or-

Select the Listener option from the File menu in the Data Explorer Client.

The following sections in the web interface correspond to the Listener main features:

♦ “Listener Section” on page 57

♦ “License Server” on page 59

♦ “Service Listener” on page 63

♦ “Inetd Listener” on page 65

♦ “Responder” on page 66

By default, all informational pages are accessible by normal users. The Administration pages require a password to access.

Note: The default user and password for administrator pages is “administrator” and no password. The Listener administrator may change this policy by changing the user and admin variables in the listener.ini file or by creating and setting user passwords in the web interface.

Listener SectionThe Listener pages display environment information about the Listener and allow an administrator to manage Listener users and passwords. The Listener page, which is the default home page for the Listener, displays the welcome message and the set of modules and their status. All modules should be in the running state, except if the license server has been disabled through the listener.ini file.

Data Explorer Listener Web Interface 57

Configuration The Configuration subsection displays the configuration environment of the Listener. These values are read from the listener.ini file.

System Information The System Information subsection displays the machine environment in which the Listener is running.

LogThe Log section displays the last page of entries in the Listener log file. Links at the top of this page display the entire log or back up log in a separate browser.

Administration The Administration subsection requires an administrator login to be viewed. This section allows you to:

♦ Stop the Listener. Stop the Listener by selecting the Stop button next to the Listener entry. Once the Listener is stopped, there is no way to start the Listener using the web interface. Operating system commands must be used to restart a stopped Listener.

♦ Manage Authentication Information. The authentication information stored on the Authentication Administration page provides access to the Listener web interface. For non-root installations on Unix-like systems, this information stored on this page also provides authentication for project access.

To access the Authentication Administration page , select the Edit button next to the Authentication Manager entry. Each user has a name, password, group membership settings, and administrator flag. On the Authentication Administration page administrators can perform the following actions:

♦ Create new users. Fill in the top box with a user name, the administrator flag, and password.

♦ Assign groups to users. This feature only applies to non-root installations on Unix-like systems. If Data Explorer is installed by a non-root user, use the groups feature to manage user authentication for Data Explorer projects. Enter group names separated by a semicolon, e.g., “Group1; Group2”. Users can access projects if they are the creator/owner or if they are a member of the group assigned to the project.

♦ Change a password. Type in the new password and click Update.

♦ Change the administrator flag. Select or clear the Administrator check box and click Update. The Administrator flag allows user to have access to user and administrator pages in the Listener web interface, while the user flag (administrator flag not checked) only allow access to user pages.

♦ Delete a user. Click Delete to remove the user. Note that you cannot delete the default administrator user.


License ServerThe License Server pages allows the user to view and control the licenses that are managed by the license server module of the Listener. The License Server page displays information gathered from the license.elf file along with current usages.

Figure 5-2 shows the License Server Information page:

Figure 5-2. License Server Information


Licenses Figure 5-3 shows the Licenses section which displays the currently allocated license tokens:

Each license token shows:

♦ Client. The client identifier internal to the Listener.

♦ Session. The session number to which the token is allocated.

♦ User. The name of the user of the application requesting the token.

♦ Host. The host on which the application requesting the token is running.

♦ UMI. The UMI of the machine running the application.

♦ Feature. The name of the feature representing the license token.

♦ TID. The internal token ID.

♦ On Since. The time when the token was requested.

The token usage is stored in the file:

Install_dir/informatica/listener/log/tokens.log

Figure 5-3. License Summary


ReportsFigure 5-4 shows the License Server Reports section which allows you to create license usage reports using the options described below:

♦ Report Type. In the Report Type section, choose one from each group of the following options:

The following options determine the type of information you can view:

♦ Feature Only. Includes features for the selected time component.

♦ Feature / Host. Includes features by host for the selected time component.

♦ Feature / User. Includes features by user for the selected time component.

The following options determine the time component:

♦ Summary. Only the summary information is collected.

♦ Month. Monthly summaries are created.

♦ Day. Daily summaries are created.

♦ Hour. Hourly summaries are created.

♦ Begin Date. The beginning date as defined by the first entry, today, or an arbitrary date.

♦ End Date. The ending date as defined by the first entry, today, or an arbitrary date.

♦ Skip rows with no data. Suppresses all rows that have no data in them.

Figure 5-4. License Server Reports


Sample ReportFigure 5-5 shows a sample report that illustrates a summary of features only since the start of the current year:

Each generated report contains the following information:

♦ Feature. The name of the feature. If Feature/Host or Feature/User was specified, the breakdown of the host or user also appears in this column.

♦ Tokens. The current maximum number of tokens for the feature.

♦ Req. The total number of requests for the feature.

♦ Issued. The total number of tokens issued for the feature.

♦ Number Denied. The number and reasons for denying a license token. They are:

− Out. The number of concurrent tokens has been exceeded.

− Exp. The feature was requested after its expiry date.

− UMI. The UMI of the client requesting the feature was not on its access control list.

− Not Lic. The feature was not licensed.

− Max In Use. The maximum number of tokens concurrently allocated for the feature.

− Total Time Used. The sum of the amount of time all tokens for the feature were in use.

Administration The Administration section allows you to forcibly return license tokens allocated to applications. By default, this section requires an administrator password.

Figure 5-5. Sample Data Explorer Listener Report


To force an application to give up an allocated token is sometimes referred to as “burying” the application token. Exercise care when burying an application token because it may cause the application to terminate.

Figure 5-6 shows the License Server Administration page:

Service ListenerThe Service Listener pages display information and control the servers that the Listener can start. The main page displays the configured servers, their version numbers, and names. It also displays the number of times each server was started.

Session The Session section displays information about active Data Explorer sessions. The information in the grid consists of the following:

♦ ID. An unique session ID.

♦ Service. The name of the application. The application can be in three states:

Figure 5-6. License Server Administration

State Indicator Description

Connected A client is actively connected to the application.

Disconnected + The application is running, but no client is connected.

Not Responding ++ The application has stopped updating its session information to the Listener and is presumed to have terminated. After the zombie period is up, these sessions are removed from the list.


♦ User. The user logged on to the application.

♦ Project. The name of the project the application is working on.

♦ Tasks. The current task that is being executed by the application, if any.

♦ % Done. The amount work done on the task.

Each session has additional detail about the application and the resources it is using which you can display by clicking the adjacent Detail button.

Figure 5-7 shows the Session Details page:

Administration The Administration section displays information about each active Data Explorer sessions and allows the administrator to terminate the session or cancel the current task:

♦ Terminating a Session. Click the Terminate button next to the desired session to send a message to the application to terminate. After 10 seconds, if the application has not terminated, it is forcibly terminated.

♦ Cancelling a task. Click the Details button next to the desired session to bring up the session details page. Click the Cancel button next to the task you want to cancel. If there is more than once task in the session, you may cancel any of them.

If one of the tasks is a batch program, cancelling it will take effect after the current job step is completed. Thus, if you want to cancel the batch program, cancel the current job step first.

Figure 5-7. Session Details


Figure 5-8 shows the Session Administration page:

Inetd ListenerThe Inetd Listener page display information about non-monitored applications that the Listener can start. The main page displays the configured applications, their version numbers, port, and names. It also displays the number of times each server was started.

Figure 5-8. Listener Session Administration


Figure 5-9 shows the Inetd Listener page:

Use the Inetd Listener to start Data Explorer Server Version 4.3 or earlier.

ResponderThe Responder page displays the port number through which the Data Explorer Listener responds when it receives broadcasts from applications looking for a Listener.

Figure 5-9. Inetd Listener


Figure 5-10 shows the Responder page:

Figure 5-10. Responder Page


Starting and Stopping the Listener

The Data Explorer Listener is started during Data Explorer Server setup. In general, you do not have to start or stop the Listener because it automatically detects configuration changes and auto-configures itself.

Note: While you must use the Windows Control Panel or the appropriate UNIX command to start the Listener, you can stop the Listener in both Windows and UNIX from the Data Explorer Listener web interface.

Start or Stop the Listener on Windows During the Data Explorer Listener installation, the install script installs the Listener as a service.

To stop or start the Listener:

1. From the Windows Control Panel, select Administrative Tools > Services.

On Windows XP Professional, this option is located in Control Panel > Performance and Maintenance.

2. In the Services dialog, select Data Explorer Listener.

3. Right-click Data Explorer Listener to stop or start the Listener.

Start or Stop the Listener on UNIXDuring the Data Explorer Listener installation process the startup (or boot) files for the UNIX server are modified so the Listener starts automatically when the server boots up. The startup files that are modified depend on the UNIX platform and can be used to start or stop the Listener. Note that you have to be root to perform these commands.

To start the Data Explorer Listener on UNIX, run the following command:

Install_dir/listener/bin/idelistenerctl start

To stop the Data Explorer Listener on UNIX, run the following command:

Install_dir/listener/bin/idelistenerctl stop


Listener Components

Table 5-2 summarizes the components necessary for the Data Explorer Listener:

Table 5-2. Data Explorer Listener Components

File Description

idelistenerctl (UNIX)ideListen.exe (Windows)

Data Explorer Listener service.

idelistenerctl startidelistenerctl stop

The shell script that starts/stops the Data Explorer Listener process. (UNIX only.)

listener.elf License file.

listener.ini Configuration file associated with the Data Explorer Listener.

listener.loglistener.log.bak

Log file associated with the Data Explorer Listener.

tokens.log Token usage log file.

Listener Components 69

C h a p t e r 6

Data Explorer Importer Configuration


♦ Overview, 72

♦ Customizing an Importer Configuration File, 73

♦ Customizing Flat-File Importers, 76

♦ Customizing Informix Configuration Files, 78

♦ Customizing Oracle Configuration Files, 79

♦ Customizing Sybase Configuration Files, 80

♦ Customizing Universal Database Configuration Files, 81

71

Overview

Users can import data into Data Explorer using any supported importer. Importer configuration files enable you to define environment variables that are either required by the importer or modify the behavior of the importer.

The environment settings in the importer configuration files override the same environment settings specified in ide_server.ini. For more information, see “Data Explorer Listener Configuration” on page 47. For information about ODBC configuration files, see “Data Explorer Repository Administration” on page 19.

Each importer runs as a separate process with its own environment. You may need to customize the environment for each importer you plan on using if the importer requires its own set of environment variables.

Note: Data Explorer servers installed on Linux can only use ODBC and flat file data importers.

72 Chapter 6: Data Explorer Importer Configuration

Customizing an Importer Configuration File

Within the importer configuration files, there is a list of common environment variables that are, by default, commented out (lines that start with the ‘#’ character are comments). To set an environment variable, first delete the comment character (#), and then fill in the appropriate value after the equal sign.

Before editing an importer configuration file, ensure that the file permissions allow you to do so. You can change the file permissions, if necessary.

On UNIX platforms, configuration files are located in:

/opt/Informatica/DataExplorer901/server901/config/DatabaseType

On Windows platforms, configuration files are located in:

Install_dir\server\config\DatabaseType

To edit a configuration file:

1. Open the configuration file in a text editor.

2. Uncomment the variable you want to set.

To uncomment a variable, delete the ‘#’ character preceding it on the line.

3. Type the new setting for the environment variable.

4. Save the file.

Configuration VariablesThe configuration file is separated into the following sections.

SystemThese are Data Explorer-defined variables which cannot be changed. This includes the header section.

Warning: Do not change the lines in the header section, as this will cause the importer to fail.

For example, the Sybase header section contains:

Name= Sybase"

Executable=idesybimp"

Type= Importer"

Log_Prefix= syb"

Version= version"

Create_Stamp= 40089355087871

Customizing an Importer Configuration File 73

EnvironmentThese variables control how the importer connects to its data source and process data. There are two types of variables. One, those that identify the data source, for example:

# SYBASE= /opt/sybase

# DSLISTEN= SYBASE

# DSQUERY= SYBASE

And two, those that help Data Explorer interpret the data from the data source. For example:

# TRIM_BLANKS= NONE

# COLLATING_SEQUENCE=defaultsyb.ecsf

# DECIMAL_CHARACTER= "."

# LIST_SEPARATOR= ","

# MONEY_CHARACTERS= "$"

# THOUSANDS_SEPARATOR= ""

Table 6-1 describes the environment variables:

Table 6-1. Environment Variables For Importers

Variable Name Description

TRIM_BLANKS The following defines the blank (white space) removal rule for reading in values. The valid rules are:- NONE. No stripping of blanks should take place.- LEADING. Remove leading blanks only.- TRAILING. Remove trailing blanks only.- BOTH. Remove leading and trailing blanks only.If not specified, "NONE" is the default.

DECIMAL_CHARACTER Defines the decimal point character.

LIST_SEPARATOR Defines the list separator character.The THOUSAND_SEPARATOR and LIST_SEPARATOR cannot be the same character.

MONEY_CHARACTERS Defines the set of currency characters to be ignored when converting numbers. The default value is “$”. The value can be changed for Eurpoean applications.

THOUSANDS_SEPARATOR Defines the thousands separator character to be ignored when converting numbers.The THOUSAND_SEPARATOR and LIST_SEPARATOR cannot be the same character.

COLLATING_SEQUENCE Defines the collating sequence used for validation tasks. Change this if your database installation uses something other than the default collation sequence.


InformaticaStandardSQLThis section defines the data sources and their SQL/translations. The special data source, “default”, represents the default data source. For most RDBMS data sources, there is only one data source/SQL translation pair. For ODBC, there can be several data source/SQL translation pairs. Note that one can use wildcard matching for specifying the data source. Here are some examples:

RedRocks*=SQLServer

*tera*=TERADATA

default=ANSI

<database>This section defines how the Data Explorer Standard SQL functions are translated into the SQL dialect of the given database. In general, the user does not need to modify the supplied defaults.

DrillDownTemplates that are used by the Drill Down dialog box to create Data Explorer Standard SQL that each importer then executes to return the results of the desired selection. For more information, see the discussion of customizing Drill Down templates in the Data Explorer User Guide.

IDE_COL_PROF_THREADS Defines the maximum number of threads to use for column profiling. The default is 1.

IDE_COL_PROF_THREADS values cannot override the licensed number of processors. Setting values above the licensed number of processors has no effect.

IDE_COL_PROF_TMP_PATH Defines the temporary directories used by the importers in column profiling. Each path should be separated by a semicolon, e.g., temp1;temp2;temp3. The paths use round-robin cyclic scheduling to improve performance. Each importer is assigned a temporary directory and sets the appropriate TMP or TMPDIR path to it. For increased performance, each path should specify a directory on a separate physical disk drive.

If present, IDE_COL_PROF_TMP_PATH overrides the ide_server.ini file setting.

Table 6-1. Environment Variables For Importers


Customizing an Importer Configuration File 75

Customizing Flat-File Importers

Data Explorer has three flat-file importer configuration files:

♦ idefile.ini. For the Data Explorer flat-file format.

♦ idedb2file.ini. For the DB2 LOAD Utility file format.

Some Data Explorer processes generate large temporary files. Determine a location that is appropriate to store these files. You will then need to customize the flat-file configuration files. For more information about editing configuration files, see “Customizing an Importer Configuration File” on page 73.

Flat-File Environment VariablesTable 6-2 describes the flat file importer environment variables:

DB2 Load Utility File Environment VariablesTable 6-3 describes the DB2 Load Utility environment variables:

Table 6-2. Flat File Importer Environment Variables


REPORT_INVALID_ROWS For the delimited files only, this parameter controls the number of invalid rows (rows that contain too few fields or too many) to report in the logs. This number should be in the range of 1...100,000. A value of '0' turns off reporting of mismatched rows.

TMPDIR Defines the default temporary directory for UNIX platforms.

TMP Defines the default temporary directory for Windows platforms.

Table 6-3. DB2 Load Utility Environment Variables


DEFAULT_DATE_LENGTH Defines the system length for the date data type. Valid range is 8 to 254. If your installation has overridden this display format, you need to change it here.

DEFAULT_TIME_LENGTH Defines the system length for the time data type. Valid range is 4 to 254. If your installation has overridden this display format, you need to change it here.

DEFAULT_TIMESTAMP_LENGTH Defines the system length for the timestamp data type. Valid range is 19 to 26. If your installation has overridden this display format, you need to change it here.

DEFAULT_CHAR_VALUE Defines the system default values for the char data type. The default value is “ “ (empty string).

DEFAULT_INTEGER_VALUE Defines the system default values for the integer data type. The default value is 0.

DEFAULT_FLOAT_VALUE Defines the system default values for float data type. The default value is 0.0.

DEFAULT_DECIMAL_VALUE Defines the system default values for decimal data type. The default value is 0.0.


DEFAULT_DATE_VALUE Defines the system default values for the date data type. The default value is currentdatetime. For each import, this value is computed once and used every time it is needed.

DEFAULT_TIME_VALUE Defines the system default values for the time data type. The default value is currentdatetime. For each import, this value is computed once and used every time it is needed.

DEFAULT_TIMESTAMP_VALUE Defines the system default values for the timestamp data type. The default value is currentdatetime. For each import, this value is computed once and used every time it is needed.

TMPDIR Defines the default temporary directory for UNIX platforms.

TMP Defines the default temporary directory for Windows platforms.

GENERATE_TRAILING_DECIMAL The following defines the behavior of generating packed/zoned decimals with no decimal digits. If set to "YES", a trailing decimal point is generated and column profiling interprets these numbers as decimals. If set to "NO", no trailing decimal point is generated and column profiling interprets these as integers.

Table 6-3. DB2 Load Utility Environment Variables


Customizing Flat-File Importers 77

Customizing Informix Configuration Files

If you are connecting to an Informix database, you must customize the configuration file ideifx.ini. For more information about editing configuration files, see “Customizing an Importer Configuration File” on page 73.

Informix Environment VariablesTable 6-4 describes the Informix importer environment variables:

For more information, see the Informix Administration Guide.

Table 6-4. Informix Importer Environment Variables


INFORMIXDIR Defines the absolute path to your Informix server installation.

DBPATH If you are doing automatic database redirection, uncomment DBPATH and replace /opt/informix/bin with the appropriate value.

INFORMIXSERVER Defines the name of your Informix server.

INFORMIXSQLHOSTS Defines the absolute path of your sqlhosts file.

ONCONFIG Defines the absolute path of your onconfig.std file.


Customizing Oracle Configuration Files

If you are connecting to an Oracle database, you must customize the corresponding file ideoraVersion.ini.

For information about editing configuration files, see “Customizing an Importer Configuration File” on page 73.

Oracle Environment VariablesTable 6-5 describes the Oracle importer environment variables:

For more information, see the Oracle Administration Guide.

Table 6-5. Oracle Importer Environment Variables


ORACLE_HOME Defines the absolute path to your Oracle server installation.

TNS_ADMIN Defines the path to the directory containing the tnsnames.ora file.

ORACLE_SID Defines the name of your Oracle database.

ORACLE_SERVER Defines the name of your server running Oracle.

DATE_FORMAT Defines how columns whose data type are dates are converted to display format. If omitted, the system default is used. The default display format is “DD-MON-YY” when Oracle is first installed, but it may be changed in the init.ora configuration file. If is recommended that you comment out this variable and set its value to your site’s display format if you wish to have the time component as part of the imported columns whose data types are dates and/or you wish to override the default display format.

NLS_LANG Specifies language-dependent behavior for a session. For example:NLS_LANG=AMERICAN.AMERICA.US7ASCIIConsult the Oracle user guide for allowed values and other national language variables.

LD_LIBRARY_PATH Defines the library path if the server runs on a Solaris or Linux machine.

SHLIB_PATH Defines the library path if the server runs on a HPUX machine.

LIBPATH Defines the library path if the server runs on a AIX machine.

PATH Defines the library path if the server runs on a Windows machine.

Customizing Oracle Configurat ion Files 79

Customizing Sybase Configuration Files

If you are connecting to a Sybase database, you must customize idesybversion.ini. For more information about editing configuration files, see “Customizing an Importer Configuration File” on page 73.

Sybase Environment VariablesTable 6-6 describes the Sybase importer environment variables:

For more information, see the Sybase Administration Guide.

Table 6-6. Sybase Importer Environment Variables


SYBASE Defines the absolute path to your Sybase server installation.

DSLISTEN Defines the name of your Sybase server.

DSQUERY Defines the name of your Sybase server.


Customizing Universal Database Configuration Files

If you are connecting to a Universal Database database, you must customize the configuration file ideudb.ini. For more information about editing configuration files, see “Customizing an Importer Configuration File” on page 73.

Table 6-7 describes the DB2 Universal Database environment variables:

For more information, see the DB2 UDB Administration Guide.

Table 6-7. DB2 Importer Environment Variables


LD_LIBRARY_PATH Solaris environment variable to define the shared library search path.

DB2INSTANCE Defines the current DB2 instance name.

Customizing Universal Database Configurat ion Files 81

C h a p t e r 7

Data Explorer Projects


♦ Overview, 84

♦ Setting Project Permissions, 86

♦ Deleting Projects, 87

♦ Compacting a Project, 88

♦ Unlocking a Project on UNIX Platforms, 89

♦ Recovering a Project, 90

♦ Ensuring Project Integrity, 91

♦ Project Compatibility, 92

83

Overview

The Data Explorer Project stores all of the information that is involved in a particular Data Explorer project. Generally, users manage their own Projects.

The Data Explorer Project is a directory that contains files and one subdirectory, reports. Data Explorer treats the Project as a file system object. You should not store user files in the Data Explorer Project.

Figure 7-1 illustrates the structure of a Data Explorer Project:

Table 7-1 describes the files and directories in a Data Explorer Project:

Table 7-2 lists the directories where reports are located:

Figure 7-1. Data Explorer Project Structure

Table 7-1. Data Explorer Project Files and Directories

Name Description

<projectname> Directory that contains the Project.

local.evk Lock file.

reports Directory containing reports, logs, and generated SQL.

object.* Project storage file.

list.* Project storage file.

Table 7-2. Data Explorer Project Report Directories

Name Description

*_DDLSQL Generated SQL DDL

*_LOGABT Batch Program Log

*_LOGCPA Cross Table Profile Log

*_LOGEXP Export IROB Log

*_LOGIMP Import IROB Log

*_LOGINF Table Structural Profile Log

*_LOGITB Import Metadata Log

ReportsObject.*

Project

List.* Local.evk

84 Chapter 7: Data Explorer Projects

You can use operating system commands to move or copy a Project as long as you keep its directory structure and files intact. Be sure to move or copy all the directories and files contained in the Project.

*_LOGLDD Load Sample Data Log

*_LOGNDX Index All Primary Keys Log

*_LOGNRM Normalization Log

*_LOGPFK Profile Foreign Key Log

*_LOGPRF Profile Columns Log

*_LOGDRD Drill Down Log

*_LOGDVL Validate Model Dependencies (table) Log

*_LOGBNF Batched Table Structural Profile Log

*_LOGORP Foreign Key Analysis Log

*_LOGRUL Rule Execution Log

*_LOGSQL Generate SQL Log

*_LOGVFK Validate Foreign Key Log

*_LOGVKY Validate Key Log

*_LOGVVM Validate Vertical Merge Log

*_LOGVJN Validate Join Log

*_LOGVRD Validate Redundancy Log

*_LOGEXP Export Data Log

*_LOGURP Update Data Explorer Repository Project Log

*_LOGRRP Export to Data Explorer Repository Project Log

*_REPDOM Domain Report

*_REPREL Table Detail Report

*_REPATT Column Detail Report

Table 7-2. Data Explorer Project Report Directories

Name Description

Overview 85

Setting Project Permissions

If you want users to be able to access a project, ensure that the project permissions allow this. For Windows, project permissions are determined by operating system file permissions. For Unix-based systems, project permissions can be set by either operating system file permissions (for root user installations) or Listener authentication (for non-root user installations).

Operating System-based PermissionsFor Data Explorer installations with operating system-based permissions (Windows installations and Unix root user installations), project access requires that the following files have read/write permissions:

♦ object.*

♦ list.*

♦ reports directory and all its files

♦ local.evk

Also, the directory that contains the Project and the reports directory must have read/write/execute permissions. Data Explorer does not support read-only Projects. Data Explorer always opens a Project for writing, so never turn off write permissions on the directory that contains the Project or you will not be able to access it.

Listener-Based PermissionsFor Unix non-root user installations, project permissions are set in the Authentication Administration page in the Listener web interface. For more information on Listener-based project permissions, see “User Privileges” on page 42.


Deleting Projects

You can delete projects from within the Data Explorer Client interface, or by deleting the project directory from Data Explorer Server.

Before deleting a project, ensure that you have permission to edit the project and that the project is not currently opened by a user.

Deleting Projects in the Data Explorer Client

To delete a project in the Data Explorer Client:

1. Log into the client with permissions sufficient for deleting the project. For more information on permissions, see “User Privileges” on page 42.

2. Select the “Open Project” option from the toolbar or the “File” menu.

3. In the Open Project dialog box, click the Delete button.

Deleting Projects from Data Explorer Server

To delete a project from the Data Explorer Server:

1. Log into the server machine with permissions sufficient for deleting the project.

2. Ensure that the project is not currently opened by a user.

3. For Windows machines, delete the project using Windows Explorer. For Unix machines, use the following command:

rm -r [ProjectName]

Delet ing Projects 87

Compacting a Project

As you work with a Project, you may need to compact the Project files.

To compact a Project:

1. Start Data Explorer as described in Data Explorer Orientation Guide.

2. If the Project you want to compact is not the current Project, open it.

3. Choose Save As from the File menu on the Main Window.

4. Data Explorer copies the Project record by record, so the free space is not copied.


Unlocking a Project on UNIX Platforms

Data Explorer uses an advisory lock on the file local.evk on UNIX-based Projects when they are open. If you are running Data Explorer server on a UNIX platform, you may occasionally need to unlock a Project.

To unlock a Project on UNIX platforms:

1. Execute the command:

% mv local.evk temp


% cp -p temp local.evk


% rm temp

Unlocking a Project on UNIX Platforms 89

Recovering a Project

If the Data Explorer server experiences a fatal error while a user is working on a Project, Data Explorer saves the Project contents in a recovery Project. This file reflects the contents of the Project at the time of the fatal error. It is named project.recovery. The Project itself is in its last saved state.

If the Data Explorer Client experiences a fatal error during a long running task, the task will complete and the server will wait for a reconnect. Restart a client and reconnect to the server. You can open a recovery Project after restarting the Data Explorer Client by choosing Open Project from the Data Explorer File menu.

If the Data Explorer Server experiences a fatal error, the recovery file may contain corrupted data.

You can attempt to determine the state of the Project by:

♦ Generating the Table Detail Report

♦ Generating SQL for all physical views in all Schemas

♦ Examining the data using the Column Summary Viewer

♦ Examining the data using the Domains Viewer

If a fatal error occurs (with the exception of a network error), contact Informatica Global Customer Support.


Ensuring Project Integrity

You can ensure the integrity of a Data Explorer Project by reserving it for Data Explorer use only. Users should not modify the Project by adding or deleting files and subdirectories that are not created by Data Explorer. However, when a Project is closed, you can use your usual backup and restore utilities on it.

Ensuring Project Integrity 91

Project Compatibility

Data Explorer Projects are not compatible across platforms. Projects created on UNIX platforms cannot be opened on Windows platforms and vice versa. However, you can move a Project from one platform to another using IROB (Intermediate Representation of Data Explorer Objects) format files.


C h a p t e r 8

Data Explorer FTM/XML


♦ System Requirements, 94

♦ Repository Connection, 95

♦ Security, 97

♦ License Tokens, 98

♦ Installing the Software, 99

♦ User Administration, 100

♦ ODBC, 101

♦ DTD Migration, 102

93

System Requirements

The following sections describe the minimum system resources required to install and run Data Explorer FTM/XML.

HardwareData Explorer FTM/XML requires the following minimum hardware configuration:

♦ 1.0 GHz Pentium III or equivalent processor (or greater)

♦ 512 MB of RAM (if you are working with extremely large Schemas, you might need additional memory)

♦ 30 MB available hard disk space

♦ Screen resolution of 1024 x 768 pixels or above (recommended)

SoftwareData Explorer FTM/XML requires:

♦ Windows

♦ User name and password to access the Data Explorer Repository

♦ Appropriate client libraries for the DBMS housing the Data Explorer Repository

♦ A web browser that supports frames, to view HTML-based map reports

94 Chapter 8: Data Explorer FTM/XML

Repository Connection

When you open Data Explorer FTM/XML, you establish a connection to the Data Explorer Repository using the Connect to Repository dialog box. The Connect to Repository dialog box allows you to configure and connect with either an ODBC driver or a JDBC Type IV Driver. Before you can connect with an ODBC driver, you must first set up ODBC data sources. For more information, see the Microsoft Windows documentation.

Note: All Type IV drivers have to be in the classpath in the Data Explorer properties.

You must have a valid Data Explorer Repository permission, user name and password, which you can obtain from your database administrator.

The Connect to Repository dialog box displays the Connection and User ID entered the last time you logged in to the Repository. The first time you open Data Explorer FTM/XML or Data Explorer Repository Navigator, these boxes are blank.

To create a new JDBC or ODBC connection, click the Configuration tab.

To configure a connection for Data Explorer FTM/XML:

1. In the Connection Name box, enter a connection name.

For a JDBC Type IV driver, enter a unique name.

For an ODBC driver, the name must match the name (DSN name) entered when you used the ODBC data source administrator function built into Windows to create the ODBC connection.

2. In the Connection Type box, specify the connection type (ODBC driver, or JDBC Type IV driver).

3. If you are using ODBC, click Save. Otherwise, follow the remaining instructions to configure your JDBC connection.

4. Select your RDBMS.

The driver name is automatically supplied in the Driver Name field.

5. In the Host field, enter the name of the host on which the RDBMS server is running.

6. In the Port field, enter the port number on the host the RDBMS uses to listen for new connections.

7. In the Database field, enter the name of the database on the RDBMS to which you want to connect.

8. The Driver Name field is automatically populated with the DataDirect Technologies driver name when you select an RDBMS, or Oracle driver (if you selected Oracle for the RDBMS).

Note: If you are using a different driver, you must install the driver in the appropriate directory as described previously, then modify the classpath.

Repository Connection 95

9. Select the Override Default URL check box if you want to automatically populate the JDBC URL field with the current JDBC information from the fields at the top of the tab.

You can then edit the JDBC URL field. If you make additional changes to the JDBC fields (such as hostname and port), the changes will not appear in the JDBC URL field until you clear the Override Default URL box, then select it again. You may still need to reenter any changes made to the JDBC URL field.

10. Verify the JDBC URL. If necessary, modify the JDBC URL.

If you selected Informix as the RDBMS, you must select the Override Default URL check box. Then, enter the Informix server name at the end of the JDBC URL field (after informixserver=).

11. Click Save.

The information becomes available on the Login tab of the Connection dialog box.

Note: To delete connection information, edit the user.properties file. Use any text editor and remove the connection information as desired. Remove the name of the connection from the Connections line. Then, remove the line with the same name as the connection. Be sure to create a backup file before editing so that mistakes can be rolled back.


Security

Data Explorer FTM/XML does not have built-in security mechanisms. Security for the Data Explorer Repository connection is implemented by requiring a user name and password to connect to the database that supports the Data Explorer Repository.

Security 97

License Tokens

The number of available Data Explorer FTM tokens (enabling licensed user access) determines the number of simultaneous sessions permitted.

Merely opening the Data Explorer FTM/XML interface does not use a token. A token is used each time a user creates or opens a Repository project (creating a new project automatically opens the new project). When a user closes a project or exits a Repository, a token is released. The time that a user has a project open is called a session.

A ten-pack license has ten tokens, so any 10 users can have projects open simultaneously. If all 10 tokens are in use, the next user who tries to open or create a new Repository project receives an error message.


Installing the Software

To install Data Explorer, use the product installation CD. For more information, see the Data Explorer Installation Guide.

Data Explorer FTM/XML defaults to install in the directory C:\Program Files\Informatica Data explorer. You can redirect the install directory.

Warning: Do not change the default install location of the Data Explorer FTM/XML help system. Doing so could cause the help system to become unusable.

The Data Explorer FTM/XML help system defaults to the same directory as the Repository Navigator and Data Explorer FTM/XML.

Installing the Software 99

User Administration

Data Explorer FTM/XML does not require special user administration actions. Licensing tokens determine how many people can run Data Explorer FTM/XML. For more information, see “License Tokens” on page 98.


ODBC

To use Data Explorer FTM/XML, the user must have a valid ODBC connection to the Data Explorer Repository. For more information, see the Data Explorer System Administration Guide.

To use the DataDirect Technologies ODBC drivers, read the Data Explorer Release Notes from the Data Explorer installation CD. You may need to follow special installation procedures depending on your database software.

ODBC 101

DTD Migration

In this version of Data Explorer FTM/XML, you can view DTDs in a hierarchical tree structure. In previous releases, you could only view DTDs in a flat structure.

This change is not backward-compatible. DTDs stored in the Data Explorer Repository using Data Explorer FTM/XML 4.1 or earlier are stored in a format that cannot be read by the current release. Attempting to use this release to work with these older DTDs can cause the program to hang. To preserve these earlier DTDs, you must migrate the DTDs to the new hierarchical format.

Note: This migration must be done before the Data Explorer Repository is updated to this release. DTDs that are not migrated before the Repository update, and all tags associated with them, will be lost. You may need to confer with your system administrator or project manager to ensure migration takes place before the Repository is updated.

Migration preserves all Notes, Rules, and Action Items associated with DTD components, but does not preserve links. Because Transformations are associated with links, Transformations are lost in migration.

DTD mapping reports show any links and Transformations associated with a DTD before migration, and can serve as a guide to make it easier to reestablish links and Transformations after migration. The report generation process always uses the same file name for the first HTML file in the report, so if you generate multiple DTD mapping reports, you must save each report in a separate folder.

DTDs are not flagged with the release level of the program, so there is no certain way to tell if a DTD was stored in the Repository by Data Explorer FTM/XML 4.1 or earlier. Duplicate component names can provide a clue to release level: 4.1 allowed creating components with duplicate names, but the current release does not. So if a DTD has components of the same type with the same name, that DTD was probably stored by release 4.1, and probably should be migrated.

If you want to keep all links and tags associated with a DTD, migration is a four-phase process:

1. Generate a DTD mapping report.

2. Generate a DTD output file.

3. Reimport the DTD from the DTD output file.

4. Optional. Recreate the links and Transformation tags.

Phase 1. Generate Mapping ReportThis phase is performed in Data Explorer FTM/XML release 4.1. For each DTD that you migrate, generate a mapping report.

Instructions for generating mapping reports are the same for 4.1 and this release.


Phase 2. Generate DTD FileThis phase is performed in Data Explorer FTM/XML release 4.1. For each DTD that you migrate, generate a DTD output file. Make sure you select ALL check boxes to preserve all tags.

Instructions for generating DTD output files are the same for 4.1 and this release.

Phase 3. Reimport DTDThis phase is performed in this version of Data Explorer FTM/XML. For each DTD that you migrate, reimport the DTD.

Phase 4. Recreate LinksAfter you reimport the DTD, you can optionally recreate any links and Transformations that had been associated with the release 4.1 DTD.

DTD Migration 103

C h a p t e r 9

Data Explorer Importer Administration


♦ Overview, 106

♦ Data Explorer Source Profiler, 107

♦ Data Explorer Import for VSAM, 108

105

Overview

This chapter briefly describes system administration issues related to Data Explorer Import for VSAM, Data Explorer Import for IMS, and Data Explorer Source Profiler. For more information about these products, see the Data Explorer Source Profiler User Guide.

106 Chapter 9: Data Explorer Importer Administration

Data Explorer Source Profiler

Data Explorer Source Profiler runs on Windows. The user interface requires that the Data Explorer Repository be installed and available to the machine running the user interface. The Repository may be installed local or remote to the machine running the user interface and is accessed using ODBC drivers. For information about Data Explorer Repository system administration issues, see “Data Explorer Repository Administration” on page 19.

Data Explorer Source Profiler requires the use of a standard Data Explorer token during the user session.

The user is responsible for:

♦ Moving the extract command file generated from the Data Explorer Source Profiler to the MVS system running the Extract for VSAM component.

♦ Moving the output from the extraction back to the Data Explorer Source Profiler user interface.

Data Explorer Source Profiler 107

Data Explorer Import for VSAM

The MVS component of Data Explorer Import for VSAM does not have any special system administration issues. The product does not require SMP for installation and runs as a typical non-APF authorized executable. The executable may be installed in a user or system load module library. The executable runs as a batch job and utilizes standard system security methods. The user submitting the job must have at least READ access to the referenced input dataset.

The Data Explorer Import for VSAM component is license-locked to a specific CPU serial number. A key is required to execute the program and may be obtained from Informatica Global Customer Support. Unlimited concurrent use is allowed on authorized CPUs.

108 Chapter 9: Data Explorer Importer Administration

A p p e n d i x A

Troubleshooting

This appendix includes the following topics:

♦ Overview, 110

♦ Data Explorer Client, 111

♦ Data Explorer Server, 113

♦ DBMS Connectivity, 114

♦ License Server, 116

♦ Resources, 118

♦ Error Reporting, 119

♦ Operating System-Specific Issues, 122

109

Overview

This appendix describes some common problems related to administering Data Explorer and solutions to these problems.

Some problems can be diagnosed with tracing. For more information, see “Error Reporting” on page 119.

110 Appendix A: Troubleshooting


DLL Access DeniedWhen trying to create a new catalog, the user gets an error message such as:

DLL access denied.

Cause: The user needs have read/execute permission on the Data Explorer server directory (and all intermediate directories), but does not have them.

Action: Grant the user read/execute permission on the Data Explorer server directory, including all intermediate directories.

Disabled Autosync FlagsAfter importing an IROB file into a catalog, the user gets an error message such as:

Disabled Autosync Flags.

Cause: The user may have exported a catalog to IROB after completing dependency profiling. The user imported the IROB file into an existing catalog and identified new dependencies and synonyms.

Action: Export the catalog to IROB, then import this IROB file into a new catalog. This resets the autosync flags.

or

Action: Transfer the file in binary mode.

Cannot Reconnect to SessionWhile running the Data Explorer Client, the user sees an error message:

A network error has occurred. Cannot reconnect to session...

Cause: This message displays if the user attempts to reconnect to a session which is no longer active. The session window in the client displays “dead” sessions if the following has occurred:

− The machine Data Explorer Server is running on crashes or is rebooted while Data Explorer server is running.

− Data Explorer Server exits abnormally without cleaning out the session file.

Action: Delete the old session file, even if the Data Explorer Server is no longer running. For more information on killing a session automatically, see the Data Explorer Orientation Guide.

Data Explorer Client 111

Manually Cleaning Out Session FilesTo clean out a session, you can use the Data Explorer Listener web interface. If the Data Explorer Server is on the local machine, type the following command in a browser to open the web interface:

http://localhost:1416

(or the port number that was selected at installation time). You can also use the IP address instead of the Data Explorer Server machine name. For example:

http://10.10.1.89:1416

Do not to remove session files that may be in use. To check for active Data Explorer Servers that use the web interface, select Session from the Service Listener section. This displays the sessions the Listener is currently managing.

To kill a session the Listener indicates is still active:

1. Select Administration from the Service Listener section and log in with administrator privileges.

2. Select the Terminate button for the session you want to kill.

3. Refresh the page and the session should be gone.

Figure A-1. Sessions in Data Explorer Listener Web Interface


Data Explorer Server

Server Connection IssuesWhile trying to initiate a Data Explorer Server session, the client is trying to start the Data Explorer Server cannot start the server. The following error message is returned:

Server ‘name’ may not exist or Data Explorer <Version Number> is not installed.

Cause: You may see this error if the Data Explorer Server has not been installed or not installed properly.

Action: Install Data Explorer as described in the Data Explorer Installation Guide.

or

Cause: The Listener configuration file does not contain an entry for a Data Explorer Server that is compatible with the client.

Action: Make sure that there is an entry in the [Services] section of the listener.ini file that points to a service that can start the correct version of the Data Explorer Server.

For example, to start Data Explorer Server 9.0.1, there should be an entry similar to the following:

[IDEServer 901]

Service= “Ide Server”

Version= 9.0.1

Type= service

Executable= /opt/informatica/DataExplorer901/server901/bin/ide_server

Arguments= “/opt/informatica/DataExplorer901/server901/bin/ide_server”

Description= “Ide Server Build 9.0.1”

or

Cause: A firewall is preventing the client from communicating with the Listener on its port.

Action: The client must be able to connect to the Listener on its well-known port of 1416. Any firewall must allow TCP/IP to connect bi-directionally to this port. Also, each Data Explorer Server that is started must likewise have a port that the client can connect to using TCP/IP. See the [FireWall] section of the listener.ini file for more information.

Data Explorer Server 113

DBMS Connectivity

Unable to Connect to DBMSData Explorer is unable to connect to a database and the user is getting an unclear error message from the DBMS.

Cause: You may not have changed the environment variables in the importer configuration file.

When connecting to RDMSs from Data Explorer, a configuration file must be set up with your site-specific information. Typically, this means setting environment variables so that Data Explorer, acting as a database client application, can successfully query the database for table names and data.

When the configuration file is improperly configured, Data Explorer will not be able to connect to your RDBMS. In some instances, the error messages returned by the RDBMS can be confusing or misleading. Error messages, such as those shown in the following, are often a sign of an incorrect configuration file:

− Informix. “Informix returns, -461: unable to load locale categories.” This message generally indicates that the INFORMIXDIR environment variable is not properly.

− Oracle. “Fatal: libclntsh.so.1.0: can't open file: error= 2 (Communication Failed).” This message generally indicates that the ORACLE_HOME environment variable is not properly set.

− Sybase. “Sybase returns, Error Code 25001, an unrecoverable error has occurred.” The SYBASE environment variable is not correctly set.

− UDB 5. “Could not execute UDB Version 5 (Communication Failed).” The DB2INSTANCE environment variable is not properly set.

Action: Edit the importer configuration file and consult your database administrator for additional help. For more information about editing the importer configuration file, see “Data Explorer Importer Configuration” on page 71.

Error While Exporting Project to RepositoryData Explorer returns the following error message when you export a project to the Repository.

Can bind a long value only for insert into a Long Column

Cause: You are attempting to export a Project from Data Explorer to an Oracle instance that contains repository tables with long varchar Column datatypes.

Action: Look for the following section in the ide_server.ini file:


# In addition, some ODBC drivers do not allow one to bind a LONG field to a non-LONG attribute. For Oracle exports, it is best to set these environment variables to avoid export errors.

# IDE_STRING_LOGIC=1

# IDE_STRING_TEXT=1

To resolve this error, uncomment the variables and restart the Data Explorer Client.

ORA-08177: can't serialize access for this transaction

Cause: There is an interference from another process in the database when you are accessing a table for inserting records.

Action: Look for the following section in the ide_server.ini file:

# Some Oracle installations return "ORA-08177: can't serialize access for this transaction" when exporting to the repository. This has to do with interference from another process in the database while we are accessing a table for inserts. To get around this, the export code has to retry the insert. To enable retrying, uncomment out the following.

# IDE_RETRIES=3

To resolve this error, uncomment the Ide_Retries=3 variable and save the ide_server.ini file.

DBMS Connectivity 115

License Server

License Server DiagnosisCause: The license server is not working and you need information to help diagnose

the problem.

Action: Check the license server log file in the appropriate directory for messages.

− On UNIX, the log file is in: Install_dir/listener/log/listener.log.

− On Windows, the log file is in: Install_dir\listener\log\listener.log.

No Licenses AvailableYou have installed a new license file, but when you start Data Explorer, the following message appears even though tokens are available:

No tokens are currently available.

Cause: The license server only allocates tokens to the licensed machines and only up to the number specified in the license.elf file. If the number of requests for a token on a given machine exceeds the number of tokens available, the request is denied, yielding this message.

Action: Make sure the machine you are using is licensed by checking its UMI against the UMI in the license.elf file. If the UMI is not in the license.elf file, use a machine that you are authorized to use. If there are not enough tokens, contact Sales to acquire more tokens.

License Server Not AvailableWhile trying to open a catalog, the user gets the following message:

Additional information: No such host ‘amber1’ Reference Code: e0000000.

Cause: If a user cannot establish a connection when trying to connect to a specific host, the license server variable IDE_ELMHOST in the ide_server.ini file may be incorrect. For example, the entry should read: IDE_ELMHOST=@amber1.

Action: Resolve the host name where the Listener that is acting as the license server is located. Also be sure to keep the @ symbol before the host name.

License Server Key File InvalidThe license server key file is invalid.

Cause: You may have edited or moved the key file. If you have done this, the key file is invalid.


Action: Reinstall the key file. If that does not work, then try reinstalling the license server. For more information, see “License Server” on page 53.

Temporary Key ProblemsYou installed a temporary key for training purposes and you cannot revert to your production key.

Cause: You may not have used the correct method for reinstalling your production key.

Action: Delete or move to an alternate location all keys from the key directory then reinstall the license server. For more information, see “License Server” on page 53.

License Server 117

Resources

Kernel ExceptionCause: You get a kernel exception.

Action: Contact Informatica Global Customer Support. You may need to send the log file.


Error Reporting

You can enable error tracing options to help troubleshoot Data Explorer Client and Server errors.

Note: Enable the options described in this section only if you are instructed to do so by Informatica Global Customer Support. Tracing requires system resources to log and communicate Data Explorer Client or Server operations, which results in a proportional decrease in system performance.


To enable tracing options for the Data Explorer Client:

1. From the menu, click Settings > Options.

The Options dialog box appears.

2. Click the Client tab.

3. Set the tracing options.

For more information on the other fields, see Chapter 3, “Data Explorer Basics” in the Data Explorer User Guide.

Data Explorer ServerYou can enable server tracing in the ide_server.ini file and in the Data Explorer Client.

The server tracing options in the ide_server.ini file and the Data Explorer Client generate different result sets. When there are problems with server or client operations, Informatica Global Customer Support may need both the result sets to isolate the problem area. If you enable server tracing in the ide_server.ini file the result set contains more details. Therefore,

Table A-1. Tracing Options for Data Explorer Client

Option Description

Select the directory where the trace files will be placed

The path where the Data Explorer Client saves the trace files. Click Browse to choose a location.Data Explorer does not generate trace files if you do not specify the location.

Trace Files to Generate - Messaging. Enables message level tracing. Data Explorer Client stores the messages sent from the client to the server in the file IDE_MessageOut.xml. It stores the reply messages from the server to client in the file IDE_MessageIn.xml. Both XML files reside on the client machine in the location you specify.- Threads. Enables more verbose client tracing. Data Explorer Client stores the entries in the file IDE_Debug.txt. This file resides on the client machine in the location you specify. It can also reside on any location on the network.

Error Reporting 119

Informatica Global Customer Support prefers that you enable server tracing both in the ide_server.ini file and in the Data Explorer Client.

To enable server tracing in the ide_server.ini file:

1. Look for the following section in the ide_server.ini file:

#################################################################

# Server Debugging - Variables for starting the log #

#################################################################

## To help diagnose server start up problems, uncomment out DEBUG_LOG_FILE and DEBUG_LOG_LEVEL variables. The debug log file may be changed to any file. In general, these should not be left uncommented to avoid multiple Ide users from overwriting the same log file.

# Windows:

# DEBUG_LOG_FILE=$(IDE_HOME)\log\ide.log

# UNIX:

# DEBUG_LOG_FILE=$(IDE_HOME)/log/ide.log

# The following defines the verbosity of the debug trace file:

# 1 - Errors only

# 2 - Verbose

# DEBUG_LOG_LEVEL=2

2. Uncomment the variables and save the ide_server.ini file.

The trace file is created in the log directory.

On UNIX, the default path is:

/opt/Informatica/DataExplorer901/server901/log

On Windows, the default path is:

Install_Directory\Server\log

The name of the trace file is:

ide.log

To enable server tracing in the Data Explorer Client:

1. From the menu, click Settings > Options.

The Options dialog box appears.

2. Click the Server tab.


3. Set the debugging options:

For more information on the other fields, see Chapter 3, “Data Explorer Basics” in the Data Explorer User Guide.

Table A-2. Debugging Options

Option Description

Debugging level The debug levels are:- None. Server error logging is disabled.- Errors Only. Server errors are written to the log file.- Verbose. Server tracing and messaging are written to the log file.

Cleanup trace files at startup Ensures that trace files are removed when you start Data Explorer. When a user first logs in to the Data Explorer Server, trace files for the user that are older than the given number of days are deleted from the log directory.

Remove trace files older than [n] days Specifies the number of days a trace file remains in the system.

Error Reporting 121

Operating System-Specific Issues

Catalogs created on Windows cannot be read directly on UNIX Platforms.

Cause: The catalog formats of Windows and UNIX systems are incompatible.

Action: To move a catalog between Windows and UNIX systems in either direction, export the catalog to an Intermediate Representation of Data Explorer Objects (IROB) file and then import it.

When your machine name and login name are the same under Windows, the login authentication process attempts to validate using the local account services rather than validating using the Domain.

Cause: Login errors occur under Windows when the local machine name and the login name are identical.

Action: To ensure correct validation, specify Domain Name\User Name in the User Name field of the login dialog box. This authenticates the user for the Domain instead of the local system.

4.0 catalogs from UNIX-based servers display incorrect column profile times in 9.0.1.

Cause: Data Explorer 4.0 catalogs from UNIX-based servers will display incorrect column profile times in Data Explorer 9.0.1.

Action: To correct this, profile the column using Data Explorer 9.0.1.


I n d e x

Aapplication servers

configuring Windows 17architecture

license server 53autosync flags

troubleshooting 111

Ccleanup scripts

Repository 26configuration files

importer 72, 73customizing

DB2 importer configuration files 81Informix importer configuration files 78Oracle importer configuration files 79Sybase importer configuration files 80

DData Explorer

changing installations 3importers 4overview 3projects 84

Source Profiler 107system requirements 9troubleshooting 110

Data Explorer Importfor VSAM 108

Data Explorer Listeneroverview 4

Data Explorer Serveroverview 3

data sourcesconfiguring Microsoft SQL Server 32DB2 Universal Database 30setting up 27

Databaseconfiguration variable 75

databasestroubleshooting connections 114

DB2 Universal Databasecustomizing importer configuration files 81customizing importer files 76data sources 30environment variables 39

DLL accesstroubleshooting 111

DrillDownconfiguration variable 75

droptables.sqlrepository cleanup script 26

123

EELF file

obtaining 54Environment

configuration variable 74environment variables

flat file importers 76for UNIX 38

Ffile systems

Windows 44flat files

customizing importers files 76FTM/XML

license tokens 98Repository connection 95requirements 94security 97

Ggroups

user and group authentication models 45

Iimporters

configuration file locations 73configuration files 72, 73customizing DB2 environment variables 76customizing flat file 76Data Explorer 4

InformaticaStandardSQLenvironment variable 75

Informixcustomizing importer configuration files 78

Llicense servers

troubleshooting 116licenses

changing 54FTM tokens 98obtaining 54server architecture 53tokens 53

troubleshooting 116Listener

changing port numbers 52Data Explorer 4overview 48starting and stopping 68

Mmanaging

swap space 15temporary space 12

Microsoft SQL Serverdata sources 32

Oobtaining

licenses 54ODBC

environment variables 39Repository requirements 21setting up 27testing connectivity 39

Oraclecustomizing importer configuration files 79

overviewRepositories 20

Ppermissions

projects 86port numbers

changing Listener 52privileges

UNIX 42Windows 43

projectscompatibility 92compressing 88ensuring integrity 91overview 84permissions 86recovering 90requirements 13unlocking on UNIX 89

124 Index

Rrecovering

projects 90Repositories

cleanup scripts 26connecting with FTM/XML 95data sources 27ODBC requirements 21overview 20

Repositorydisk space requirements 22

requirementsData Explorer FTM/XML 94project 13

Ssecurity

FTM/XML 97UNIX 42Windows platforms 43

ServerData Explorer 3

serverstroubleshooting connections 113

sessionstroubleshooting connectivity 111

setting upODBC 27

Source Profileroverview 107

startingListener 68

stoppingListener 68

swap filesmanaging on UNIX 15managing on Windows 15

swap spacemanaging 15

Sybasecustomizing importer configuration files 80

Systemconfiguration variable 73

system requirementsData Explorer 9

Ttemporary space

managing 12tokens

license 53tracing

performing 119troubleshooting

Data Explorer 110

UUMASK

UNIX 42UMI

definition 55UNIX

connectivity 38environment variables 38managing swap files 15privileges 42security 42unlocking projects 89UNMASK 42

usersuser and group authentication models 45

VVSAM

Data Explorer Import 108

WWindows

configuring application servers 17file systems 44managing swap files 15privileges 43

Index 125

126 Index

System Administration Guide - Informatica · Informatica Data Explorer Standard Edition System...

Documents

Transcript of System Administration Guide - Informatica · Informatica Data Explorer Standard Edition System...