IBM DB2 UNIVERSAL DATABASE
UDB 
Fixpak 16
 
 
(C) COPYRIGHT INTERNATIONAL BUSINESS MACHINES CORPORATION 2001.
    ALL RIGHTS RESERVED.
 
 
  
 
 
|  DB2 Universal Database Fixpak 16!
| Fixpak 16 
 
  1.0  ɱ
| 1.1  FixPak 6 Τ٭ع
| 1.2  Fixpak 16 Τ٭ع
  1.3  ܨΤ٭ݴջ
| 1.4  DB2 ů FixPak ǵ
| 1.5  Web ̱
  1.6  DB2 ˷ع
 
  2.0  ADMINISTRATION GUIDE (ŧ)
  2.1  Implementing Your Design
|   2.1.1  Creating a Database/Creating a Summary Table
      2.1.1.1  Isolation Levels and Query Optimization Using Summary Tables
       or Replicated Tables
|     2.1.1.2  REFRESH IMMEDIATE
    2.1.2  Optimizing Performance When Data is Placed on RAID Devices
| 2.2  Controlling Database Access
|   2.2.1  DRDA-only Client Authentication
|   2.2.2  Change Password Enablement for EEE
|   2.2.3  Encrypted Passwords
|   2.2.4  Loginretries Option (AIX only)
  2.3  Auditing DB2 Activities
  2.4  Utilities for Moving Data
    2.4.1  LOAD, IMPORT, and EXPORT File Formats/PC/IXF File Formats/Code
     Page Considerations
|   2.4.2  Overview of the LOAD Process (Pending States)
    2.4.3  Load Restriction on Windows NT and OS/2 SMP-enabled Machines
|   2.4.4  Error Scenario for the Autoloader Program on DB2 UDB Enterprise
|    - Extended Edition for Windows NT
|   2.4.5  Changes in Autoloader Authentication Behavior
|   2.4.6  Import/Export of WSF Data Containing BIGINT
  2.5  Recovering a Database
    2.5.1  Backing Up and Restoring Databases on Different Operating
     Systems
    2.5.2  Considerations for Using ADSTAR Distributed Storage Manager
     (ADSM)
|   2.5.3  Configuration Parameters for Database Logging
|   2.5.4  Frequency of Backups and Time Required
    2.5.5  Recovering a Database (Planning to Use the ROLLFORWARD Command)
| 2.6  Using DB2 with an XA-Compliant Transaction Manager
| 2.7  Application Considerations
| 2.8  Understanding the SQL Compiler
  2.9  Operational Performance
|   2.9.1  Managing the Database Buffer Pool
    2.9.2  Managing Multiple Database Buffer Pools
| 2.10  Using the Governor
  2.11  Configuring DB2
|   2.11.1  Encrypted Passwords
    2.11.2  Tuning Configuration Parameters/Optimizing Hash Join
     Performance
|   2.11.3  Database Log Files
|   2.11.4  Trusted Clients and DRDA Use
|   2.11.5  Loginretries Option (AIX only)
    2.11.6  ASLHEAPSZ and Failed Queries Against Large Tables
    2.11.7  INTRA_PARALLEL Setting on an SMP Machine (Windows NT)
| 2.12  HACMP ES for AIX
| 2.13  DB2 Registry and Environment Variables
  2.14  Additional Rah (Run All Hosts) Information (Solaris and AIX only)
| 2.15  Unicode/UCS-2 and UTF-8 Support in DB2 UDB
|   2.15.1  Introduction
|     2.15.1.1  UTF-8
|   2.15.2  UCS-2/UTF-8 Implementation in DB2 UDB
|     2.15.2.1  Code Page/CCSID Numbers
|     2.15.2.2  Creating a UCS-2 Database
|     2.15.2.3  Data Types
|     2.15.2.4  Identifiers
|     2.15.2.5  UCS-2 Literals
|     2.15.2.6  Pattern Matching in a UCS-2 Database
|     2.15.2.7  IMPORT/EXPORT/LOAD Considerations
|     2.15.2.8  Incompatibilities
| 2.16  Using Virtual Interface (VI) Architecture
|   2.16.1  Overview of Virtual Interface Architecture
|   2.16.2  Install DB2 Universal Database Version 5.2 (EEE)
|   2.16.3  Running DB2 UDB V5.2 for Windows NT with GigaNet Interconnect
|     2.16.3.1  Setup Procedure for GigaNet Interconnect
|     2.16.3.2  Implement DB2 to Run Using VI
|   2.16.4  Running DB2 UDB V5.2 for Windows NT with ServerNet Interconnect
|     2.16.4.1  Setup Procedure for ServerNet Interconnect
|     2.16.4.2  Implement DB2 to Run Using VI
 
  3.0  API REFERENCE (ŧ)
  3.1  sqlecrea - Create Database
  3.2  sqlubkp - Backup Database
  3.3  sqlugtpi - Get Table Partitioning Information
  3.4  sqluhget - Retrieve DDL Information From the History File (New API)
  3.5  sqluhgne - Get Next Recovery History File Entry
  3.6  sqluhops - Open Recovery History File Scan
  3.7  sqluload - Load
  3.8  SQLE-CLIENT-INFO (Connection Settings)
  3.9  SQLFUPD
  3.10  SQLUHINFO (SQLUHADM)
  3.11  sqluvput - Writing Data to Device
 
  4.0    WINDOWS ů OS/2 Ȣ
| 4.1  WINDOWS NT ĸ DB2START ȴع
  4.2   WINDOWS NT  WINDOWS 95  SQL Ȣ (IBM
       VISUALAGE FOR COBOL)
  4.3   JAVA Ȣ APPLETS
    4.3.1   MICROSOFT SDK FOR JAVA  WINDOWS 
|   4.3.2  ڷ׻ůΤȢ AND SQLJ Щ SCRIPT
 
  5.0   UNIX Ȣ
  5.1  SCO UnixWare 7
|   5.1.1  ʩ̪
    5.1.2  MICRO FOCUS COBOL 
  5.2   Java Ȣů Applets
    5.2.1  躡HP-UX, SCO UnixWare 7
|   5.2.2  CLOSE()  SILICON GRAPHICS IRIX ̱
    5.2.3  ڷ׻ůΤȢ AND SQLJ Щ SCRIPT
 
  6.0  CLI GUIDE AND REFERENCE (ŧ)
  6.1  SQLPREPARE()
  6.2  SQLDISCONNECT()
  6.3  Using Stored Procedures - Programming Stored Procedures To Return
   Result Sets
  6.4  Stored Procedure Catalog Tables
| 6.5  SQLFetchScroll(SQL_FETCH_RELATIVE, -1)
 
  7.0  COMMAND REFERENCE (ŧ)
  7.1  db2cmd - Open DB2 Command Window
  7.2  db2eva - Event Analyzer
  7.3  db2evmon - Event Monitor Productivity Tool
  7.4  db2look - DB2 Statistics Extraction Tool
  7.5  db2set - DB2 Profile Registry Command
  7.6  db2upd52 - Update Catalog to Support Version 5.2
  7.7  BACKUP DATABASE
  7.8  EXPORT
  7.9  GET DATABASE MANAGER CONFIGURATION
    7.9.1  New Parameter catalog_noauth
|   7.9.2  TRUST_ALLCLNTS Updated
  7.10  IMPORT
  7.11  LIST DATABASE DIRECTORY
  7.12  LOAD
  7.13  RECONCILE
  7.14  RESTORE DATABASE
| 7.15  START DATABASE MANAGER
 
  8.0  EMBEDDED SQL PROGRAMMING GUIDE (ŧ)
  8.1  Writing User-Defined Functions (Interface between DB2 and a UDF/The
   Arguments Passed from DB2 to a UDF)
  8.2  Programming in C and C++ (Programming Considerations/Host Variables)
    8.2.1  Declaring Host Variables
    8.2.2  Handling Graphic Host Variables (Windows Operating System)
| 8.3  Programming in COBOL (Host Structure Support)
  8.4  Programming in Java (Creating Java UDFs and Stored
   Procedures/Classes for Java Stored Procedures and UDFs)
  8.5  A JDK Problem That Affects JDBC Programs Running under a DBCS Code
   Page on AIX
  8.6  JDBC Sample Applications Restriction on SCO UnixWare 7
| 8.7  JDBC 2.0
| 8.8  Perl Access to DB2 UDB Databases
  8.9  DATALINK Data Type
  8.10  Multithreaded UNIX Applications Working with Code Page and Country
   Code
  8.11  Error Processing for UDF Call Types
    8.11.1  Scratchpad and Final Call Type Arguments for EXTERNAL TABLE
     Functions
|   8.11.2  External UDF Error Processing Model for Scalar Functions
| 8.12  NOT ATOMIC Compound SQL
  8.13  Other Revisions to the Book
    8.13.1  Revisions to Chapter 1
    8.13.2  Revisions to Chapter 4
    8.13.3  Revisions to Chapter 7
    8.13.4  Revisions to Chapter 9
    8.13.5  Revisions to Chapter 15
    8.13.6  Revisions to Sample Programs
    8.13.7  Other File Changes
 
  9.0   DB2 ڷ׻
  9.1   CCA ڷ׻̯ DB2 Τ
 
  10.0  عٶ
  10.1  DBI1768W 
  10.2  SQL0270N /̧
  10.3  SQL0351N 
  10.4  SQL0352N 
  10.5  SQL0357N 
  10.6  SQL0432N 
  10.7  SQL0680N 
  10.8  SQL0903N ̧
  10.9  SQL0911N ̧
  10.10  SQL1339N 
  10.11  ܢ SQL5048N  SQL0901N
| 10.12  ع SQL5156N: DBM ٶ TRUST_ALLCLNTS ԫ
  10.13  SQL20053N ̧
  10.14  SQL20056N ع
  10.15  SQL30083N ̧
 
  11.0  ʸī
  11.1   WINDOWS 95  NEC PC98 ĸ DB2
  11.2   WINDOWS 95  NEC PC98 ĸ DB2
  11.3  
    11.3.1  /Τܨ
    11.3.2   OS/2 ڷ׻ڷ׻ܨ/
|   11.3.3  DBCS  OS/2 WARP SERVER V.4 ̱
    11.3.4  DB2 FOR AS/400  DB2 UNIVERSAL DATABASE FOR AIXOS/2
             WINDOWS NT  SNA 
|   11.3.5  ġԶĸTP  DB2 ġ
    11.3.6  ǿ۾ǵ/DB2 FOR OS/390 ԺȢ
  11.4  DB2 UNIVERSAL DATABASE ī ()
  11.5  DB2 UNIVERSAL DATABASE ī ( AIX ĸϯ)
  11.6   (ջš޴ٴΤ̯ DB2
  EXTENDED ENTERPRISE EDITION FOR WINDOWS NT/
  ϯ ID ) 
| 11.7  NT ĸ DB2START ȴع
  11.8  ڷȹ
| 11.9  ڷ DB2 ȹ 5.0 Ū̯ȹ 5.2
| 11.10  LOTUS DOMINO GO WEB Τ DB2 ٯȢ
  11.11  Ȣ ( JAVA Ȣ)
  11.12  DB2 ܨթ ()
 
  12.0  DB2 File Manager ʸī
| 12.1  ڷ׻īů
| 12.2  DB2 DATA LINKS MANAGER  DB2 ENTERPRISE -
|  Extended Edition ˷ĸ
  12.3  DLFM ܡϯ ID
  12.4   NFS ̽ DLFS ˷
  12.5   DATALINKS ڷ׻̱
  12.6  DLFM 
  12.7  ܡȢΤĸڷߦ
| 12.8  ެǹ (DL_NUM_BACKUP)
  12.9  DB2 File Manager ߦ٭
  12.10  ڷݨܡ (CLP) ǡԺ
 
  13.0   5.2 Щشʸīܡ
  13.1  DB2 ܨ
  13.2  
    13.2.1  OS/2 
      13.2.1.1  ƫڷ׻̯Τ
    13.2.2  NetFinity Ȣ - Windows NT
    13.2.3  Communications Server for Windows NT - ԺȢ
    13.2.4  š޴ջϯ Client Application Enabler Ū
    13.2.5   Windows ODBC ϯ Client Application Enabler Ū
    13.2.6  AIX ĸ ADSM ԺȢȹ
    13.2.7  CS/NT  Windows NT ĸ̥
      13.2.7.1  ش CS/NT ɢҳ Windows NT Τ٭Ȣ
    13.2.8  WINDOWS NT ů WINDOWS 95 ĸ DCE 
|   13.2.9   WINDOWS NT ȹ
    13.2.10  SYSADM  (UNIX)
  13.3  ̧ɢ
    13.3.1   CS/NT ῡܡȢ(SPM)
      13.3.1.1   SNA SPM ع
|   13.3.2   AIX  4.1.4 ů 4.1.5 ĸ NIS ϯ
    13.3.3  db2cpic.dll - WINDOWS NT ů WINDOWS 95 ĸ SNA 
            Ȣ
    13.3.4   Windows NT  DB2NTMEMSIZE 
  13.4  ̱ SNAPSHOT Ȣ
  13.5  ڷ Web ̽ DB2 
  13.6   RS/6000 SP ĸٶ
  13.7  š޴ջ CCA ̱ (DB2 UNIVERSAL
        DATABASE EXTENDED ENTERPRISE EDITION FOR AIX)
  13.8  ڷ (UNIX)
  13.9  DB2 SNMP ܡȢ
  13.10   5.2 Щشʸīܡ
  13.11  ܡ OS/2 ϯܡ
  13.12   SOLARIS ĸ EEE  ADSM ڷ׻ -  db2profile
  13.13  ڷ DB2 FOR OS/390  5 ̯ DB2 UNIVERSAL DATABASE  5
  13.14   WINDOWS NT Τĸ NETBIOS
  13.15  ش - Windows NT  Windows 95
  13.16   DB2 JDBC APPLET Τɢ WINDOWS NT Τ٭Ȣ
  13.17  ̽ڵع
    13.17.1  ĸ
    13.17.2  DB2 
    13.17.3   PostScript 
    13.17.4  ĸ
    13.17.5  ǵΤ
    13.17.6  ĸ
  13.18  Microsoft Internet Explorer 4.0 HTTP 1.1 
  13.19  ̱ڷܡ
 
  14.0  REPLICATION вٶ
  14.1  ƺ
  14.2  ع
  14.3  DB2 UDB  5.2 ̽Ȣ LONG VARCHAR
| 14.4  DataJoiner Replication Administration Tool README
 
  15.0  DB2 Ȣв
  15.1   Application Developer's Kit
 
  16.0  SQL REFERENCE (ŧ)
| 16.1  Naming Conventions (cursor-name)
| 16.2  ALTER TABLE
| 16.3  Compound SQL
  16.4  CREATE FUNCTION
    16.4.1  DBINFO
|   16.4.2  Error Processing for UDF Call Types
| 16.5  CREATE TABLE
|   16.5.1  CREATE TABLE Description
|   16.5.2  CREATE TABLE Notes
| 16.6  CREATE TABLESPACE
  16.7  Appendix A. SQL Limits/Database Manager Limits
 
  17.0  SYSTEM MONITOR GUIDE AND REFERENCE (ŧ)
  17.1  Event Monitors are Created with Local Scope
 
  18.0  
| 18.1   (/֪)
| 18.2   JAVA Ȣů APPLETS 
  18.3  WEB ̱ NETQUESTION ( WEB ̱ǵ)
  18.4  ̱̱
| 18.5  Web ̱ JAVA DEVELOPMENT KIT ȹ
 
  19.0  OS/2  HTML Τ¡ HTML ǵ
  19.1   OS/2 ĸ HTML ˷
    19.1.1  OS/2 ĸ˷ǿ۾ǵ
    19.1.2   OS/2 ĸ TCP/IP
    19.1.3   TCP/IP 
    19.1.4  ˷
    19.1.5  Τ
    19.1.6  Τ
  19.2   HTML ˷
    19.2.1  ٯΤ
    19.2.2  Ţ
    19.2.3  ̯
    19.2.4  Ϊ ( 500 )
    19.2.5  ٯΤ٭Ȣ̯
  19.3   HTML ˷
    19.3.1   OS/2 ĸ˷
 
  20.0  WINDOWS NTWINDOWS 95 ů WINDOWS 98  HTML Τ¡
         HTML ǵ
  20.1  ̱
  20.2   WINDOWS NTWINDOWS 95 WINDOWS 98 ĸ HTML ˷
    20.2.1  ˷ǿ۾ǵ
    20.2.2  ǿЩ˷
    20.2.3  ˷
    20.2.4  ҳΤ̧
    20.2.5   LAN ĸ DB2 UNIVERSAL DATABASE
    20.2.6  Τ
  20.3   HTML ˷
    20.3.1  ٯΤ
    20.3.2   NETSCAPE  INTERNET EXPLORER  PROXIES
    20.3.3  ĸ WINDOWS 95 
    20.3.4  Ϊ ( 500 )
    20.3.5  ̯
    20.3.6  ٯΤ٭Ȣ̯
  20.4  20.4   HTML ˷
    20.4.1  Ϊ˷ڷ WINDOWS NTWINDOWS 95  WINDOWS 98
            
    20.4.2  ܨѥ
 
| 21.0  NETQUESTION 
| 21.1   UNIX ƺĸ IMNSEARCH (NETQUESTION) 
| 21.2   HPUX-10 ĸ NetQuestion
| 21.3  UNIX ƺ  NETQUESTION ů DB2 CAE
| 21.4  AIX ƺĸ NETQUESTION DBCS
 
  22.0  SCOHELP ع
  22.1  ̽ SCOhelp
  22.2  ե SCOHELP
    22.2.1  ֱ
    22.2.2  Τĸ SCOHELP 
    22.2.3   SCOHELP ڷ׻
    22.2.4  
  22.3  WEB ̱ SCOHELP
  22.4  Support Level Supplement (SLS) PTF7004A, the UnixWare 7 Language
   Extension Supplement
    22.4.1  ԺŻ LANG.CONF
  22.5  SCOhelp ſ
 
 
|  DB2 Universal Database Fixpak 16
 
  ެ⡡  ҳס
 
|  README ȴ DB2  Fixpak 16 ȹع
| ԶعԺ DB2  5.2 ȹ README 
| ȩ
 
| ١Ժ۾ "|" ġȹĸڷ 5.2
| README ȹѥԺ
 
  Ķܨع
 
|      IBM DB2 Universal Database Enterprise Edition,  5.2
|      IBM DB2 Universal Database Enterprise - Extended Edition,  5.2
|      IBM DB2 Universal Database Personal Edition,  5.2
|      IBM DB2 Universal Database Workgroup Edition,  5.2
|      IBM DB2 Personal Developer's Edition,  5.2
|      IBM DB2 Universal Developer's Edition,  5.2
 
  ެ⡡  VisualAge for Basic  Lotus Approach ܨ
        ܵ
 
  ĶܨħԶɱ (ҳ READDCS.TXT)

 
       IBM DB2 Connect Enterprise Edition 5.2
       IBM DB2 Connect Personal Edition 5.2
 
| DB2 Universal Database ջ 5.2 仡
  ԯġ
 
  o   ܡī
| o    Windows ů OS/2 Ȣ
  o   CLI Guide and Reference
  o   Embedded SQL Programming Guide
  o   Master Index
  o   SQL ī
  o   System Monitor Guide and Reference
  o   Troubleshooting Guide
 
| ȴ 5  5.2  DB2 ڰů 5.2 ڰ
|  (ɱ 5.2 ) 
| աġ㡤 DB2 Universal Database  5 ů DB2 Connect  5
| FixPaks ƫݷҳ 5.2 ڰ䡤
 
| ȴ DB2 ջѥѡ
| աů README عɻƫ̽ڵ DB2 ع
 
 
| Fixpak 16 
 
|  Fixpak 16 աش
 
| ƫ EEE ĸ
 
| DB2 ܡϯܢƫ̧ĩش "DB2CHGPWD_EEE=<boolean>"
|  AIX ů NT EEE ˷ĸعٶ 2.2.2,
| "Change Password Enablement for EEE"
 
|  DRDA ڷ׻š
 
| DB2 UDB Τܢƫٴš DRDA ڷ׻ DB2 for MVS ů OS/390
| DB2 for VM ů VSEDB2 for OS/400ů̧ڷ׻ѡ
| ƫ̧ڷ׻Ķ衤ĸڷ׻ƫ̿غꧡ
| ڷ׻šܢľƫɲڷ׻š
| عٶ 2.2.1, "DRDA-only Client
| Authentication"
 
| 
 
| ٶ "AUTHENTICATION" ջܡȢԫ "SERVER_ENCRYPT"
| "SERVER_ENCRYPT" ܢ̦ȴ "SERVER" ̦ȴ
| ̡ع ٶ 2.2.3, "Encrypted
| Passwords"
 
| JDBC 2.0
 
|  Fixpak  JDBC 2.0 ٯȢ JDBC 2.0  API
ԡҳ
| DB2 Engine ݽ֪ȴȴ
| Ρعٶ 8.7, "JDBC 2.0"
 
| NETQUESTION 
 
| ݡ
 
| o    UNIX ƺĸ IMNSearch (NetQuestion) 
 
| o    HPUX-10 ĸ NetQuestion
 
| o   UNIX ƺ NetQuestion ů DB2 CAE
 
| o   AIX ƺĸ NetQuestion DBCS
 
| عٶ 21.0, "NetQuestion "
 
| 
 
| ڰɢȩΡREFRESH IMMEDIATE ܢƫħ
| عٶ 135  16.2, "ALTER TABLE"ů
| 16.5, "CREATE TABLE"
 
| UNICODE/UCS-2 ů UTF-8 
 
| Unicode ġǹ
| ȴҤŧISO ů IEC 10646  (ISO/IEC 10646) 
| ̦ȴ 2  (UCS-2) ů 4  (UCS-4)  Multiple-Octet
| Coded Character Set (UCS)DB2 UDB  UCS-2ǩɻȴܡ
| Unicode
 
| ȴħ UCS-2  Unicode 䡢ASCII ů̱ľ̥Զۡ
| Щҳ 0 ASCII Ȣů ASCII ˷
| ҳ UCS-2 㡢 NULL ƫܢǶ
| ɡ ASCII Ȣơġ UTF-8 
| ƫ̱ԶDB2  UTF-8
 
| عٶ 2.15, "Unicode/UCS-2 and UTF-8 Support
| in DB2 UDB"
 
|  (VI) 
 
| DB2 UDB ܢ̥Զ VI 硨ġԶ GigaNetƶġԶ
| ServerNetعٶ 2.16.3, "Running DB2 UDB V5.2
| for Windows NT with GigaNet Interconnect"
 
 
  1.0  ֱ
 
 
| 1.1  FIXPAK 6 Τ٭ع
 
| DB2 UDB  5.2 ƫůҡľƫɢҳġԶ FixPak  DB2
| UDB  5.0 ȹĸƫ 5.0 Ūҳ̦ȴ FixPak 6 
| 5.2ɢ˷ FixPak 6  PTF Ķ㡨
 
  +---------------------------------------------------------------------------+
| |  1. DB2 UDB FixPak 6  PTF                                         |
  +--------------------------------------+-------------------------------------+
| | ɢ˷                             | DB2 UDB FIXPAK 6  PTF         |
  +--------------------------------------+-------------------------------------+
| | AIX                                  | U459852                             |
  +--------------------------------------+-------------------------------------+
| | HP-UX 10                             | U459854                             |
  +--------------------------------------+-------------------------------------+
| | HP-UX 11                             | U459872                             |
  +--------------------------------------+-------------------------------------+
| | OS/2                                 | WR09073                             |
  +--------------------------------------+-------------------------------------+
| | SCO UnixWare 7                       | U459870                            |
  +--------------------------------------+-------------------------------------+
| | Silicon Graphics IRIX                | U459871                             |
  +--------------------------------------+-------------------------------------+
| | Solaris                              | U459853                             |
  +--------------------------------------+-------------------------------------+
| | Windows 95Windows 98 ů Windows    | WR09074                             |
| | NT                                   |                                     |
  +--------------------------------------+-------------------------------------+
| | Windows 3.1 ů Windows for           | WR09076                             |
| | Workgroups 3.11                      |                                     |
  +--------------------------------------+-------------------------------------+
 
  ެ⡡   AIX  4.2 ĸDB2 File Manager PTF U456886
        ޲ԫ
 
  Ķ SCO  PTF  DB2 UDB  5.2  SCO UnixWare 7
  ĸƫ
 
       ptf7003c
       ptf7010
       ptf7013
 
| ȴ DB2 UDB  5.2  SCO PTF ޲عƫ 1.3, "Product
| and Service Technical Library" ̯
 
   Sun Cluster ĶԺȢȹ
 
       105786-05     - Sun Cluster 2.x  Solaris 2.6 ĸ
                       ԺȢ
       2.1_105458-06 - DB2  Sun Cluster ԺȢȹ
 
 
| 1.2  Fixpak 16 Τ٭ع
 
|  FixPak ܢȴ DB2 UDB  5.0 Ȣȹĸ

| ŷ DB2 UDB  5.2 ȹҳħ̯ 5.2 ȹƫĶ
| ů FixPak 6ġ FixPak 6ƫſ Fixpak 16
 
  +--------------------------------------------------------------------------+
| |  2. DB2 UDB Fixpak 16  PTF                                        |
  +-------------------------------------+-------------------------------------+
| | ɢ˷                            | DB2 UDB Fixpak 16  PTF        |
 +-------------------------------------+-------------------------------------+
| | AIX                                 | U478404                             |
  +-------------------------------------+-------------------------------------+
| | HP-UX 10                            | U478406                             |
  +-------------------------------------+-------------------------------------+
| | HP-UX 11                            | U478407                             |
  +-------------------------------------+-------------------------------------+
| | OS/2                                | WR21262                             |
  +-------------------------------------+-------------------------------------+
| | SCO UnixWare 7                      | U478408                             |
  +-------------------------------------+-------------------------------------+
| | Silicon Graphics IRIX               | U478409                             |
  +-------------------------------------+-------------------------------------+
| | Solaris                             | U478405                             |
  +-------------------------------------+-------------------------------------+
| | Windows 95, Windows 98, and Windows | WR21263                             |
| | NT                                  |                                     |
  +-------------------------------------+-------------------------------------+
| | Windows 3.1 and Windows for         | WR21264                             |
| | Workgroups 3.11                     |                                     |
  +-------------------------------------+-------------------------------------+
 
 
 
  1.3  ܨΤ٭ݴջ
 
  ޲ DB2 ع (Ⱥ README ѥܢ)
  Ķ World Wide Web ĸ Product and Service Technical
  Library
 
       http://www.software.ibm.com/data/db2/library
 
 
| 1.4  DB2 ů FIXPAK ǵ
 
|  5.2 㡢ȩܨǹ仡DB2 ƫĸ
|  HTML ֪Ȣ̽ڵǹľƫ PDF ů postscript ֪Ȣ̽ڵ
| ٶĸ 1.3, "Product and Service Technical Library"̽ڵ
| ĸ̽ٶաġ
| ůƫ SBOF-8921-00
|  5.2 ޲ġ
 
  ǿЩī read1297.txtread0398.txt read0698.txt 
   FixPak ع㡨
 
  o   ǵ㡢Ϊ
 
  o   աġΪ仡
 
  Ⱥعǩ㡤
 
 
| 1.5  WEB ̱
 
  Web ̱ Java  DB2 Universal Database Control
  Center (ġջܡ̦) DB2 ̱
  ҳΡƫ̡̦ҳܡ躡 Web ̱
  ҳ DB2 JDBC  Java applet
 
| FixPak ҡWeb ̱סŪܢۡΪȴ
| Web ̱סڷĶ URL Ķơɻƫ̽ڵ
 
       http://www.software.ibm.com/data/db2/udb/webcc
 
 
  1.6  DB2 ˷ع
 
   DB2 ˷ܨ޲عƫ "DB2 Family Newsletter"
  ()ӼȴǶ塢 1-905-316-4733Ķ
  ع
 
       
       Ƴ
       
       
       Щ DB2 ܨ
 
 
  2.0  ADMINISTRATION GUIDE (ŧ)
 
 
  2.1  IMPLEMENTING YOUR DESIGN
 
 
| 2.1.1  CREATING A DATABASE/CREATING A SUMMARY TABLE
 
 
  2.1.1.1  Isolation Levels and Query Optimization Using Summary Tables or
  Replicated Tables
 
  In general, a summary table or a
  replicated table is used for optimization of a query if the isolation level
  of the summary table or the replicated table is higher than or equal to the
  isolation level of the query.  For example, if
  a query is running under the cursor stability (CS) isolation level, only
  summary tables and replicated tables that are defined under CS or higher
  isolation levels are used for optimization.
 
 
| 2.1.1.2  REFRESH IMMEDIATE
 
| The REFRESH IMMEDIATE option is now supported for summary tables. For more
| information, please see section 16.2, "ALTER TABLE", and section
| 16.5, "CREATE TABLE".
 
 
  2.1.2  OPTIMIZING PERFORMANCE WHEN DATA IS PLACED ON RAID DEVICES
 
  This section describes how to optimize performance when data is placed on
  Redundant Array of Independent Disks (RAID) devices. In general, you should
  do the following for each table space that uses a RAID device:
 
  o   Define a single container for the table space (using the RAID device).
 
  o   Make the EXTENTSIZE of the table space equal to, or a multiple of, the
      RAID stripe size.
 
  o   Ensure that the PREFETCHSIZE of the table space is:
 
      -   the RAID stripe size multiplied by the number of RAID parallel
          devices (or a whole multiple of this product), and
 
      -   a multiple of the EXTENTSIZE.
 
  o   Use the DB2_PARALLEL_IO registry variable (described below) to enable
      parallel I/O for the table space
 
  o   Use the DB2_STRIPED_CONTAINERS registry variable (described below) to
      ensure extent boundaries are aligned in the table space.
 
  DB2_PARALLEL_IO
 
  When reading data from, or writing data to table space containers, DB2 may
  use parallel I/O if the number of containers in the database is greater than
  1. However, there are situations when it would be beneficial to have parallel
  I/O enabled for single container table spaces. For example, if the container
  is created on a single RAID device that is composed of more than one physical
  disk, you may want to issue parallel read and write calls.
 
  To force parallel I/O for a table space that has a single container, you can
  use the DB2_PARALLEL_IO registry variable. This variable can be set to "*"
  (asterisk), meaning every table space, or it can be set to a list of table
  space IDs separated by commas. For example:
 
       db2set DB2_PARALLEL_IO=*        {turn parallel I/O on for all table spaces}
       db2set DB2_PARALLEL_IO=1,2,4,8  {turn parallel I/O on for table spaces 1, 2, 4, and 8}
 
 
  After setting the registry variable, DB2 must be stopped (db2stop), and then
  restarted (db2start), for the changes to take effect.
 
  DB2_STRIPED_CONTAINERS
 
  Currently when creating a DMS table space container (device or file), a
  one-page tag is stored at the beginning of the container. The remaining pages
  are available for data storage by DB2, and are grouped into extent-sized
  blocks.
 
  When using RAID devices for table space containers, it is suggested that the
  table space be created with an extent size that is equal to, or a multiple
  of, the RAID stripe size. However, because of the one-page container tag, the
  extents will not line up with the RAID stripes, and it may be necessary
  during an I/O request to access more physical disks than would be optimal.
 
  DMS table space containers can now be created in such a way that the tag
  exists in its own (full) extent. This avoids the problem described above, but
  it requires an extra extent of overhead within the container. To create
  containers in this fashion, you must set the DB2 registry variable
  DB2_STRIPED_CONTAINERS to "ON", and then stop and restart your instance:
 
       db2set DB2_STRIPED_CONTAINERS=ON
       db2stop
       db2start
 
  Any DMS container that is created (with CREATE TABLESPACE or ALTER
  TABLESPACE) will have new containers with tags taking up a full extent.
  Existing containers will remain unchanged.
 
  To stop creating containers with this attribute, reset the variable, and then
  stop and restart your instance:
 
       db2set DB2_STRIPED_CONTAINERS=
       db2stop
       db2start
 
  The Control Center and the LIST TABLESPACE CONTAINERS command will not show
  whether a container has been created as striped or not. They will continue to
  use "file" or "device", depending on how the container was created. To verify
  that a container was created as striped, you can use the /DTSF option of
  DB2DART to dump table space and container information, and look at the type
  field for the container in question. Also, the query container APIs,
  sqlbftcq( ) and sqlbtcq( ), can be used to create a simple application that
  will display the type.
 
  Definitions for these new types have been added to the sqlutil.h header file:
 
       #define SQLB_CONT_STRIPED_DISK 5      /* DMS: Striped disk */
       #define SQLB_CONT_STRIPED_FILE 6      /* DMS: Striped file */
 
 
| 2.2  CONTROLLING DATABASE ACCESS
 
 
| 2.2.1  DRDA-ONLY CLIENT AUTHENTICATION
 
| In the section "Selecting an Authentication Method for Your Server", two new
| paragraphs are added:
 
| To protect against all clients except DRDA clients from DB2 for MVS and
| OS/390, DB2 for VM and VSE, and DB2 for OS/400, set the "trust_allclnts"
| parameter to DRDAONLY. Only these clients can be trusted to perform
| client-side authentication. All other clients must provide a user ID and
| password to be authenticated by the server.
 
| The "trust_clntauth" parameter is used to determine where the above clients
| are authenticated: if "trust_clntauth" is "client", authentication takes
| place at the client. If "trust_clntauth" is "server", authentication takes
| place at the client when no password is provided and at the server when a
| password is provided.
 
 
| 2.2.2  CHANGE PASSWORD ENABLEMENT FOR EEE
 
| DB2 Administrators may now allow others to change passwords on AIX and
| Windows NT EEE systems through the profile registry variable
| "DB2CHGPWD_EEE=<boolean>".
 
| NOTE:  This is not supported on Sun EEE systems.
 
| The default for this variable is "NOT SET" (disabled). Other values for
| "DB2CHGPWD_EEE" are the standard boolean values used by other DB2 profile
| variables.
 
| The DB2 Administrator is responsible for ensuring that the passwords for all
| nodes are maintained centrally using either a Windows NT Domain Controller on
| Windows NT, or NIS on AIX.
 
| NOTE:  If the passwords are not maintained centrally, enabling the
|        "DB2CHGPWD_EEE" variable will allow for the possibility that passwords
|        may not be consistent across all nodes. That is, if a user uses the
|        "change password" feature, then the user's password will only be
|        changed at the node to which they connect.
 
 
| 2.2.3  ENCRYPTED PASSWORDS
 
| A new "DBM CFG" (database manager configuration) value "SERVER_ENCRYPT" has
| been added for the parameter "AUTHENTICATION". "SERVER_ENCRYPT" now has the
| same rules as "SERVER" authentication, with the added feature that any
| passwords sent over a network will be encrypted. Also, a new value
| "DCS_ENCRYPT" has been added for DB2 Connect support. This value is
| documented in the file READDCS.TXT, available with the following products:
 
|      IBM DB2 Connect Enterprise Edition, Version 5.2
|      IBM DB2 Connect Personal Edition, Version 5.2
 
| The same parameter has been added to the "CATALOG DATABASE" authentication
| clause as an option.
 
| RESTRICTIONS
 
| 1.  In terms of database support, only DB2/390 and DB2 UDB accept an
|     encrypted password for this initial implementation. DB2 UDB support is
|     initially limited to the following platforms:
 
|     o   Windows 95/98
 
|     o   Windows NT
 
|     o   AIX
 
| 2.  No CCA/Discovery Support for the new ENCRYPT versions of authentication
|     is supported.
 
| 3.  No Support in the Control Center.
 
| 4.  No encryption of the new password for change password support.
 
| 5.  No DCE Directory support of the new authentication types SERVER_ENCRYPT
|     and DCS_ENCRYPT.
 
| 6.  No DRDA AS support for encryption.
 
| 7.  Any encryption which occurs will only be for the password - user ID and
|     the remainder of the flow remain unencrypted.
 
 
| 2.2.4  LOGINRETRIES OPTION (AIX ONLY)
 
| DB2 Universal Database on AIX has added the functionality to log failed
| password attempts with the operating system and detect when a client has
| exceeded the number of allowable login tries as specified by the
| "loginretries" parameter.
 
 
  2.3  AUDITING DB2 ACTIVITIES
 
  There is no audit support for summary tables, typed tables, or replicated
  tables.
 
 
  2.4  UTILITIES FOR MOVING DATA
 
 
  2.4.1  LOAD, IMPORT, AND EXPORT FILE FORMATS/PC/IXF FILE FORMATS/CODE PAGE
  CONSIDERATIONS
 
  If you want to load or import an IXF format file containing DBCS data, you
  must have the corresponding conversion files (located in sqllib\conv)
  installed on your client machine. The names of these conversion files contain
  both the source and the target code page numbers; the extension is always
  converting code page 932 to 943.
 
  If your client machine does not have the appropriate conversion files, you
  can copy them from a server machine to the sqllib\conv directory of the
  client machine. Be sure to copy the files from a compatible platform; for
  example, if the client is running on a UNIX based operating system, copy the
  files from a server that is also running on a UNIX based operating system.
 
 
| 2.4.2  OVERVIEW OF THE LOAD PROCESS (PENDING STATES)
 
| In the section "Overview of the LOAD Process", a revision to the paragraph
| discussing pending states is made:
 
| After the LOAD process completes, if:
 
| o   "logretain" or "userexit" is "on", COPY NO (which is the default) is
|     selected and NONRECOVERABLE is not selected; or,
 
| o   "logretain" or "userexit" is "off," COPY YES is selected and
|     NONRECOVERABLE is not selected
 
| then any associated table spaces are placed in a BACKUP PENDING state.
 
| In the section "LOAD Exception Table," two new paragraphs are added:
 
| LOAD, when using the FOR EXCEPTION option, will check for rows that have
| unique index violations. LOAD will not check for constraint or foreign key
| violations other than for unique constraint violations.
 
| A "unique key" is a key that is constrained so that no two of its values are
| equal. The mechanism used to enforce the constraint is called a unique index.
| A primary key is a special case of a unique key. A table cannot have more
| than one primary key.
 
 
  2.4.3  LOAD RESTRICTION ON WINDOWS NT AND OS/2 SMP-ENABLED MACHINES
 
  In DB2 UDB Version 5.2 for Windows NT and DB2 UDB Version 5.2 for OS/2, users
  operating in an SMP environment might occasionally experience a system hang
  during a load operation. This is a known problem that occurs only during
  index creation (the BUILD phase of the load utility) involving large volumes
  of data (on the order of several gigabytes). To avoid this problem, it is
  recommended that existing indexes on a loading table be dropped before
  issuing the LOAD command, and recreated using the CREATE INDEX statement
  after the load operation has completed successfully. An alternative that does
  not require the dropping of indexes is to perform the load operation into a
  staging table that does not have any indexes defined on it. The target table
  can subsequently be updated by issuing an INSERT statement encompassing all
  of the data from the staging table.
 
 
| 2.4.4  ERROR SCENARIO FOR THE AUTOLOADER PROGRAM ON DB2 UDB ENTERPRISE -
| EXTENDED EDITION FOR WINDOWS NT
 
| The following applies to an error scenario for the autoloader program on IBM
| DB2 Universal Database Enterprise - Extended Edition for Windows NT. It
| explains the symptom and nature of the problem, and also a workaround.
 
| When running the autoloader program, "db2atld.exe", on a multi-homed machine,
| that is, a machine with multiple network cards installed, ensure that the
| machine is configured correctly by trying the following operations:
 
| o   When you type the "hostname" command on the machine where the autoloader
|     is started, and also ping this host name from the same machine, the IP
|     address returned should be the same as the result when this host name is
|     pinged from another machine in your DB2 MPP node list. If the machine is
|     not configured correctly, you will see an "SQL6555N" error from the
|     autoloader program, and you will also see the error message, "errno =
|     10061", i.e. connection refused, in the "db2diag.log" file(s) on some of
|     the loading nodes defined by the "OUTPUTNODES" parameter in your
|     autoloader configuration file.
 
| On Windows NT machines, the IP address returned for a local host name is not
| retrieved from the "DNS" or "hosts" file, but from information configured
| locally in the Control Panel network icon.
 
| There is a bug in Windows NT 4.0 where the binding order configured in the
| Control Panel network icon does not influence the IP address order returned
| on a multi-homed machine. See Microsoft Support Online article ID Q171320 to
| resolve this problem. (When you ping a host name, it is the first IP address
| that is returned.)
 
 
| 2.4.5  CHANGES IN AUTOLOADER AUTHENTICATION BEHAVIOR
 
| The concept of a local database connection has been extended for the MPP
| environments to include connections from any node of a given MPP instance.
| That means that even though the instance is configured using AUTHENTICATION
| server, a password will no longer be required if a connection is being
| attempted from one of the nodes defined in the "db2nodes.cfg" file.  The
| Autoloader will make use of this new connection behavior when the
| AUTHENTICATION flag in the "autloader.cfg" file is not set or is set to NO,
| and the password for autoloader is not defined.
 
| With this new database connection behavior, the password is only mandatory
| for autoloader execution if a password is required for remote execution of
| programs in your system. For example, if the ".rhosts" file on a UNIX system
| was not set up properly to enable "rsh" execution, then the password is
| required.
 
 
| 2.4.6  IMPORT/EXPORT OF WSF DATA CONTAINING BIGINT
 
| When exporting "BIGINT" or "DECIMAL" columns to WSF (Work Sheet Format), only
| values that fall in the range of type DOUBLE can be exported accurately.
| Values that do not fall in this range will still be exported to the target
| media. However, loading back these values (via "LOAD" or "IMPORT") may result
| in incorrect data. The end result will vary by platform.
 
 
  2.5  RECOVERING A DATABASE
 
 
  2.5.1  BACKING UP AND RESTORING DATABASES ON DIFFERENT OPERATING SYSTEMS
 
  You cannot back up a database or table space on one operating system and
  restore it to another operating system by using the BACKUP DATABASE and
  RESTORE DATABASE commands. Instead, you move tables from one operating system
  to another by using the db2move command (or by exporting, then importing or
  loading the data). For more information, see "Moving Data Using the db2move
  Tool" in the Administration Guide.
 
 
  2.5.2  CONSIDERATIONS FOR USING ADSTAR DISTRIBUTED STORAGE MANAGER (ADSM)
 
  The current ADSM client on the Windows operating system and OS/2 is
  non-reentrant, and so multiple sessions cannot be created with
  the backup, restore, or load utilities from a single machine.
 
  In a single node configuration, if a user attempts to issue a backup command
  such as:
 
       db2 backup db sample use adsm open 3 sessions
 
  DB2 will detect that multiple sessions are not supported by ADSM, and will return
  SQL2032N. The equivalent scenario also applies to load copies using ADSM.
 
  However, in an MLN configuration on Windows NT, DB2 may not be able to detect
  the use of multiple sessions on a single machine if each logical node attempts
  to create only one session. If multiple logical nodes are being backed up,
  restored, or loaded in parallel using ADSM, DB2 will allow the operation to proceed
  if each node attempts to use a single session, even though the logical nodes actually
  reside on the same physical hardware. This can lead to failed backup attempts,
  and hung load processes, and should not be attempted.
 
 
| 2.5.3  CONFIGURATION PARAMETERS FOR DATABASE LOGGING
 
| The following information should be added:
 
| The total log file size limit is 4 GB. That is, the number of logfiles
| (LOGPRIMARY + LOGSECOND) multiplied by the size of each logfile in bytes
| (LOGFILSIZ * 4096) must be less than 4 GB.
 
 
| 2.5.4  FREQUENCY OF BACKUPS AND TIME REQUIRED
 
| In the section on "Frequency of Backups and Time Required," the following
| information should be added:
 
| If the amount of time needed to apply archived logs when recovering and
| rolling forward a database is a major concern, then you will need to consider
| the cost of having more frequent backups. If your database is very active,
| more logging occurs and therefore more frequent database backups are
| recommended. More frequent database backups reduces the number of archived
| logs you need to apply when rolling forward through archived logs.
 
 
  2.5.5  RECOVERING A DATABASE (PLANNING TO USE THE ROLLFORWARD COMMAND)
 
  You cannot roll forward a partitioned database from a Version 2 client.
 
 
| 2.6  USING DB2 WITH AN XA-COMPLIANT TRANSACTION MANAGER
 
| Within the sub-section 'Using the DB2 Universal Database XA Switch' under the
| 'XA Function Supported' section in this chapter, the last line in "OS/2
| Platform" should be changed to:
 
|      With either method, you must link your application with db2api.lib.
 
  The last line in the "Windows NT Platform" section should be changed to:
 
       With any of these methods, ensure that you link with db2api.lib.
 
| References to "db2app.lib" should be changed to "db2api.lib."
 
 
| 2.7  APPLICATION CONSIDERATIONS
 
| In the section on "Locking," just before the  "Attributes of Locks" section,
| the following paragraph should be added:
 
| You may want to consider defining a monitor that will record when deadlocks
| occur. Use the CREATE EVENT MONITOR statement described in the SQL Reference
| to create the monitor.
 
| In the section on "Using a select-statement," the OPTIMIZE FOR and FETCH
| FIRST example should be modified to show the following:
 
|   SELECT EMPNAME, SALARY FROM EMPLOYEE
|      ORDER BY SALARY DESC
|      FETCH FIRST 100 ROWS ONLY
|      OPTIMIZE FOR 20 ROWS
 
| The change is simply the reversal of the order of the last two lines of the
| example.
 
 
| 2.8  UNDERSTANDING THE SQL COMPILER
 
| In the section on "Hash Join," the following information should be added:
 
| To realize the performance benefits of hash join, it may be necessary to
| change the value of the "sortheap" database configuration parameter, and the
| "sheapthres" database manager configuration parameter.
 
| For decision support queries, hash join access plans use more sort heap space
| than do non-hash join plans. When "sheapthres" is set to be very close to
| "sortheap" (that is, less than a factor of two or three per concurrent
| query), a hash join runs with much less memory than the optimizer
| anticipated. When executing with limited memory, hash joins can be very slow.
| The problem occurs in queries with multiple sorts and hash joins, in which
| the first sorts or hash joins acquire most of the available memory.
 
| The solution is to configure "sheapthres" to be large enough (relative to
| "sortheap").
 
 
  2.9  OPERATIONAL PERFORMANCE
 
 
| 2.9.1  MANAGING THE DATABASE BUFFER POOL
 
| The following information should be added:
 
| The storage associated with the buffer pool is allocated when a database is
| activated or when the first application connects to the database.
| Applications are the primary beneficiaries of the buffer pool; once
| applications are all disconnected, the storage associated with the buffer
| pool is de-allocated.
 
 
  2.9.2  MANAGING MULTIPLE DATABASE BUFFER POOLS
 
  When working on your database design, you may have determined that tables
  with 8KB page sizes were best. As a result, you should create a buffer pool
  with an 8KB page size (along with one or more table spaces with the same page
  size).
 
  If you have chosen to place your tables in an 8KB table space, it is
  recommended that your temporary table spaces have an 8KB page size as well,
  and they should be assigned to the same 8KB page size buffer pool. The
  default buffer pool (IBMDEFAULTBP) or any other 4KB buffer pool to which any
  4KB temporary table spaces are assigned should be kept small, since the
  database manager will favor the 8KB temporary table spaces, and therefore the
  8KB buffer pool.
 
 
| 2.10  USING THE GOVERNOR
 
| In the section on "The Governor Deamon" add the following information:
 
| The note concerned with using the governor as an alternate means to adjust
| agent priorities should have the following information added: "(This note
| does not apply to OS/2 or Windows NT platforms.)"
 
| In the section on the "Governor Configuration File", replace the first and
| second paragraphs of the information under "action/priority" with the
| following:
 
| Specifies a change to the priority of agents working for the application.
| Valid values are from -20 to +20.
 
 
  2.11  CONFIGURING DB2
 
 
| 2.11.1  ENCRYPTED PASSWORDS
 
| The AUTHENTICATION type (authentication) parameter is updated as follows:
 
| A new value, SERVER_ENCRYPT, has been added to this parameter. This value has
| the same rules as SERVER authentication, with the added feature that any
| passwords sent over a network are encrypted. For restrictions on the use of
| this value, see section 2.2.3, "Encrypted Passwords".
 
| A new value, DCS_ENCRYPT, has been added to this parameter. This value has
| been added for support of DB2 Connect. Details on this new value are
| documented in the READDCS.TXT file for the following two products:
 
| o   IBM DB2 Connect Enterprise Edition, Version 5.2
 
| o   IBM DB2 Connect Personal Edition, Version 5.2.
 
 
  2.11.2  TUNING CONFIGURATION PARAMETERS/OPTIMIZING HASH JOIN PERFORMANCE
 
  To realize the performance benefits of hash join, it may be necessary to
  change the value of the SORTHEAP database configuration parameter, and the
  SHEAPTHRES database manager configuration parameter.
 
  Hash join plans use more sort heap space than other join plans. When
  SHEAPTHRES is set to be very close to SORTHEAP (less than a factor of 2 or 3
  per concurrent query), hash join runs with much less memory than the
  optimizer anticipated. When executing with limited memory, hash join can be
  very slow.  The problem occurs in queries with multiple sorts and hash joins,
  in which the first sorts or hash joins acquire most of the available memory.
 
  The solution is to configure SHEAPTHRES to be large enough (relative to
  SORTHEAP).
 
  For more information about hash join, including how to determine whether a
  hash join strategy has been selected, see the description of db2expln in the
  SQL Reference, and the What's New book (Appendix A - System Monitor Guide and
  Reference Updates). See also the description of the new hash join registry
  variable (DB2_HASH_JOIN) below.
 
 
| 2.11.3  DATABASE LOG FILES
 
| The "Size of Log Files (logfilsz)" parameter is updated as follows:
 
| The total log file size limit is 4 GB. That is, the number of logfiles
| (LOGPRIMARY + LOGSECOND) multiplied by the size of each logfile in bytes
| (LOGFILSIZ * 4096) must be less than 4 GB.
 
 
| 2.11.4  TRUSTED CLIENTS AND DRDA USE
 
| The section of the chapter dealing with "Instance Administration," the "Trust
| All Clients (trust_allclnts)" parameter is updated as follows:
 
| The Range of values for the parameter now includes DRDAONLY in addition to NO
| and YES.
 
| The description for the parameter now includes the following two paragraphs:
 
| Setting the "trust_allclnts" parameter to DRDAONLY protects against all
| clients except DRDA clients from DB2 for MVS and OS/390, DB2 for VM and VSE,
| and DB2 for OS/400. Only these clients can be trusted to perform client-side
| authentication. All other clients must provide a user ID and password to be
| authenticated by the server.
 
| The "trust_clntauth" parameter is used to determine where the above clients
| are authenticated: if "trust_clntauth" is "client", authentication takes
| place at the client. If "trust_clntauth" is "server", authentication takes
| place at the client when no password is provided and at the server when a
| password is provided.
 
 
| 2.11.5  LOGINRETRIES OPTION (AIX ONLY)
 
| DB2 Universal Database on AIX has added the functionality to log failed
| password attempts with the operating system and detect when a client has
| exceeded the number of allowable login tries as specified by the
| "loginretries" parameter.
 
 
  2.11.6  ASLHEAPSZ AND FAILED QUERIES AGAINST LARGE TABLES
 
  DB2 can fail while fetching query results from large tables (that is, tables
  with greater than a 4005-character record length and more than 500 columns).
  This can be overcome by increasing the value of the ASLHEAPSZ database
  manager configuration parameter from the default value of 15.
 
 
  2.11.7  INTRA_PARALLEL SETTING ON AN SMP MACHINE (WINDOWS NT)
 
  On Windows NT, on an SMP machine, the INTRA_PARALLEL database manager
  configuration parameter will be set to YES when an instance is created with
  DB2 Version 5.2. With Version 5.0, it would have been set to NO.
 
 
| 2.12  HACMP ES FOR AIX
 
| The section "Cluster Configuration," has an example showing how to create
| containers for a two-node system that is updated as follows:
 
|      CREATE TABLESPACE TS3 MANAGED BY SYSTEM USING
|         ('/TS/cont $N%2, '/TS3/cont $N%2+2')
 
| The following containers would be used:
 
|      /TS3/cont0   - on Node 0
|      /TS3/cont2   - on Node 0
|      /TS3/cont1   - on Node 1
|      /TS3/cont3   - on Node 1
 
| The change was made to be consistent with the 'TS3' subdirectory.
 
 
| 2.13  DB2 REGISTRY AND ENVIRONMENT VARIABLES
 
 
|      Parameter       Operating         Values                    Description
|                      System
|      -------------------------------------------------------------------------------------
|      DB2CHGPWD_EEE   AIX and           Default=NULL              DB2 Administrators can allow non-
|                      Windows NT        Values: YES or NO         administrators to change passwords.
|                      (not supported                              The administrator is responsible for
|                       on Sun EEE)                                ensuring that the passwords for all
|                                                                  database partitions (nodes) are
|                                                                  maintained centrally using either
|                                                                  a Windows NT Domain Controller on
|                                                                  Windows NT, or NIS on AIX. If not
|                                                                  maintained centrally, passwords may
|                                                                  not be consistent across all database
|                                                                  partitions. This could result in a
|                                                                  password only being changed at the
|                                                                  database partition to which the user
|                                                                  connects to make the change.
 
|      Parameter       Operating      Values                       Description
|                      System
|      -------------------------------------------------------------------------------------
|      DB2_GRP_LOOKUP  Windows NT     DEFAULT=null                 Specifies which Windows NT
|                                     Values: local, domain        security mechanism will be
|                                                                  used to enumerate the groups
|                                                                  that a user belongs to.
 
| NOTE:  For FixPak 7, the "db2_grp_lookup" registry variable has been enhanced
|        to support the keyword "domain", in addition to the currently
|        supported keyword "local".
 
|             db2set db2_grp_lookup=domain
 
|        should be set under the following conditions:
 
|        1.  Your machine is not a domain-controller.
 
|        2.  Your machine is in a domain other than the domain where the domain
|            accounts reside. This occurs using the Master Account Domain model
|            where accounts are defined in a domain trusted by the resource
|            domains.
 
|        3.  You have backup domain controllers for the accounts domain.
 
|        This parameter applies to both client and server configurations.
|        Setting this parameter will tell DB2 to use a domain controller in the
|        current domain to locate a controller for the accounts domain.
 
|        PROBLEM BEING SOLVED:
 
|        In order to enumerate groups (and to find out whether you are an
|        administrator), DB2 uses an NT API to find a domain controller for the
|        domain in which the account is defined. It uses an API that tries to
|        find the Primary Domain Controller (PDC) and, failing that, a backup
|        domain controller. If the machine that is running this API is also a
|        domain controller, this will always work. If your machine is not a
|        domain controller, then this methodology will fail when the PDC is
|        down. When "db2_grp_lookup=domain", DB2 will find a domain controller
|        in YOUR domain with which to run the API to determine the domain
|        controller for the accounts domain. This will not fail when the PDC is
|        down.
 
       Parameter       Operating      Values                    Description
                       System
       ------------------------------------------------------------------------------------
       DB2_HASH_JOIN   All            DEFAULT=NO                Specifies hash join as
                                      Values: YES or NO         a possible join method when
                                                                compiling an access plan.
 
|      Parameter            Operating     Values                Description
|                           System
|      ------------------------------------------------------------------------------------
|      DB2_NEW_CORR_SQ_FF   All           DEFAULT=OFF           Affects the selectivity value
|                                         Values: ON or OFF     computed by the SQL optimizer
|                                                               for certain subquery predicates
|                                                               when it is set to ON.  It can be
|                                                               used to improve the accuracy of
|                                                               the selectivity value of equality
|                                                               subquery predicates that use the
|                                                               MIN or MAX aggregate function in
|                                                               the select list of the subquery.
|                                                               For example:
 
|                                                                 SELECT * FROM T  WHERE T.COL =
|                                                                  (SELECT MIN(T.COL) FROM T WHERE ...)
 
|      Parameter       Operating      Values                    Description
|                      System
|      ------------------------------------------------------------------------------------
|      DB2MEMDISCLAIM  AIX            DEFAULT=(not set)        Depending on the workload being executed
|                                     Values: YES or NO        and the pool agents configuration, you
|                                                              may run into a situation where the
|                                                              committed memory for each DB2 agent will
|                                                              stay above 32MB even when the agent is
|                                                              idle. This behavior is expected and
|                                                              usually results in good performance as
|                                                              the memory is kept for fast re-use.
|                                                              However, on a memory constrained system,
|                                                              this may not be a desirable side effect.
|                                                              To avoid this condition issue the
|                                                              following:
 
|                                                                 db2set DB2MEMDISCLAIM=yes
 
|                                                              Disclaiming memory tells the AIX operating
|                                                              system to stop paging the area so that it
|                                                              no longer occupies any real storage.
|                                                              Setting DB2MEMDISCLAIM to YES tells DB2
|                                                              UDB to disclaim some or all memory once
|                                                              freed, depending on DB2MEMMAXFREE. This
|                                                              will ensure that the memory is made
|                                                              readily available for other processes as
|                                                              soon as it is freed. See also
|                                                              DB2MEMMAXFREE.
 
|      Parameter       Operating      Values                    Description
|                      System
|      -------------------------------------------------------------------------------------
|      DB2MEMMAXFREE   AIX            DEFAULT=(not set)         Specifies the amount of free memory that
|                                     Values: 4000000           is retained by each DB2 agent. You may
|                                        to 256000000           set this variable to a value between 4
|                                                               and 256 MB. We recommend that if you use
|                                                               this feature, you specify a value of 8MB:
 
|                                                                  db2set DB2MEMMAXFREE=8000000
 
|                                                               See also DB2MEMDISCLAIM.
 
| There is a new registry variable to tune bufferpools. This is how you use it:
 
| 1.  Create a file
 
| 2.  Set the file as follows:
 
|          db2set DB2BPVARS=<filename>
 
| For example, the file "f:\BPVARSFILE" has the following content:
 
  -----------------------------------------------------------------------------
 
|   # turn on scatter read for raw DMS devices
|   NT_SCATTER_DMSDEVICE=1
 
  -----------------------------------------------------------------------------
 
| You would do the following using this file:
 
|      db2set DB2BPVARS=f:\BPVARSFILE
 
| The currently supported parameter is:
 
|      Parameter            Operating      Values               Description
|                           System
|      -------------------------------------------------------------------------------------
|      NT_SCATTER_DMSDEVICE  NT            DEFAULT=0(OFF)       Can be used to turn scatter
|                                          Values=0(OFF)        read on for DMS device
|                                            or 1(ON)           containers. This can only
|                                                               be enabled (turned ON) if
|                                                               DB2NTNOCACHE is set to ON in
|                                                               the registry. A warning message
|                                                               will be written to the
|                                                               db2diag.log if DB2NTNOCACHE
|                                                               is OFF (or not set), and
|                                                               scatter read will remain
|                                                               disabled. It is recommended for
|                                                               systems with a large amount of
|                                                               sequential prefetching against
|                                                               RAW containers.
 
 
  2.14  ADDITIONAL RAH (RUN ALL HOSTS) INFORMATION (SOLARIS AND AIX ONLY)
 
  The following is to be added to the rah information in Appendix Q.
 
  To enhance performance, rah has been extended to use tree_logic on large
  systems. That is, rah will check how many nodes the list contains, and if
  that number exceeds a threshold value, it constructs a subset of the list and
  sends a recursive invocation of itself to those nodes. At those nodes, the
  recursively invoked rah follows the same logic until the list is small enough
  to follow the standard logic (now the "leaf-of-tree" logic) of sending the
  command to all nodes on the list. The threshold can be specified by
  environment variable RAHTREETHRESH, or defaults to 15.
 
  In the case of a multiple-logical-node-per-physical-node system, db2_all will
  favor sending the recursive invocation to distinct physical nodes, which will
  then rsh to other logical nodes on the same physical node, thus also reducing
  inter-physical-node traffic. (This point applies only to db2_all, not rah,
  since rah always sends only to distinct physical nodes.)
 
  This version of rah has nearly identical syntax and semantics as the old
  version (supplied in the product as rah.sh_old), except for some minor
  restrictions on what options can be used:
 
  o   The user must use ksh (kornshell) as the shell. If the user is using a
      different shell, rah_tree will issue a warning and use non-tree logic.
 
  o   The single quotation mark (') prefix character, requesting echo of the
      command, cannot be honored, and is ignored.
 
  o   When both the < (all-but-me) and > (substitute <> by host index) options
      are specified, the host index is different from what non-tree rah/db2_all
      would have substituted.
 
      NOTE:  The () and ## substitutions should work identically to the old
             rah/db2_all.
 
  o   It is strongly recommended that the user ID setup at all nodes in the
      list be identical; for example, the current working directory from which
      the rah_tree command or the db2_tree command is issued should exist on
      all nodes; the rah_tree executable must be found in the current $PATH on
      all nodes (the one in effect when the rah_tree command is issued); the
      rahwaitfor executable must be found in that path, and so on. Certain
      environment differences between nodes can be tolerated, but many can not.
 
  o   The command to be executed must not start with the characters "-o, -b,
      -d, or -x", because rah_tree will interpret these to be flags.
 
  o   When specifying parallel execution, the order in which hosts return their
      output is likely to be different from the order generated by non-tree
      rah, which tends to return output in list order.
 
  o   Whenever rah_tree or db2_tree is invoked, it compares the number of
      destination nodes with a threshold value, as described above. The
      threshold can be specified by environment variable:
 
           export RAHTREETHRESH=nn
              where nn can be any positive integer
 
      or defaults to 15.
 
 
| 2.15  UNICODE/UCS-2 AND UTF-8 SUPPORT IN DB2 UDB
 
| These two standards are documented here.
 
 
| 2.15.1  INTRODUCTION
 
| The Unicode character encoding standard is a fixed-length, character encoding
| scheme that includes characters from almost all the living languages of the
| world. Unicode characters are usually shown as "U+xxxx" where xxxx is the
| hexadecimal code of the character.
 
| Each character is 16 bits (2 bytes) wide, regardless of the language. While
| the resulting 65 536 code elements are sufficient for encoding most of the
| characters of the major languages of the world, the Unicode standard also
| provides an extension mechanism, that allows for encoding as many as a
| million more characters. This extension reserves a range of code values
| (U+D800 to U+D8FF, known as "surrogates") for encoding some 32-bit characters
| as two successive code elements.
 
| The International Organization for Standardization (ISO) and International
| Electrotechnical Commission (IEC) 10646 standard (ISO/IEC 10646) specifies
| the Universal Multiple-Octet Coded Character Set (UCS) that has a 2-byte
| version (UCS-2) and a 4-byte version (UCS-4). The 2-byte version of this ISO
| standard UCS-2 is identical to Unicode without surrogates. ISO 10646 also
| defines an extension technique, for encoding some UCS-4 codes in a UCS-2
| encoded string.  This extension called UTF-16, is identical to Unicode with
| surrogates.
 
| DB2 UDB supports UCS-2, that is, Unicode without surrogates.
 
 
| 2.15.1.1  UTF-8
 
| With UCS-2 or Unicode encoding, ASCII and control characters are also two
| bytes long, and the lead byte is zero. For example, NULL is U+0000 and
| CAPITAL LETTER A is represented by U+0041. This could be a major problem for
| ASCII-based applications and ASCII file systems, because in a UCS-2 strings,
| extraneous NULLs may appear anywhere in the string. A transformation
| algorithm, known as UTF-8, can be used to circumvent this problem for
| programs that rely on ASCII code being invariant.
 
| UTF-8 (UCS Transformation Format 8), is an algorithmic transformation which
| transforms fixed-length UCS-2 and UCS-4 characters into variable-length byte
| strings. In UTF-8, ASCII characters are represented by their usual
| single-byte codes, but non-ASCII characters in UCS-2 become two or three
| bytes long. In other words, UTF-8 transforms UCS-2 characters to a multi-byte
| codeset, for which ASCII is invariant. The number of bytes for each UCS-2
| character in UTF-8 format can be determined from the following table:
 
|       UCS-2 (hex)     UTF-8 (binary)                 Description
|       ------------    --------------------------     ----------------
|       0000 to 007F    0xxxxxxx                       ASCII
|       0080 to 07FF    110xxxxx 10xxxxxx              up to U+07FF
|       0800 to FFFF    1110xxxx 10xxxxxx 10xxxxxx     other UCS-2
 
|       NOTE: The range D800 to DFFF is to be excluded from treatment
|             by the third row of this table which corresponds to the
|             UCS-4 range 0000 0800 to 0000 FFFF.
 
| In all the above, a series of x's indicate the UCS bit representation of the
| character. For example, U0080 transforms into 11000010 10000000.
 
 
| 2.15.2  UCS-2/UTF-8 IMPLEMENTATION IN DB2 UDB
 
 
| 2.15.2.1  Code Page/CCSID Numbers
 
| Within IBM, the UCS-2 code page has been registered as code page 1200. All
| code pages are defined with growing character sets, that is, when new
| characters are added to a code page, the code page number does not change.
| Code page 1200 always refers to the current version of Unicode/UCS-2, and has
| been used for UCS-2 support in DB2 UDB.
 
| A specific repertoire of the UCS standard, as defined by Unicode 2.0 and
| ISO/IEC 10646-1, has also been registered within IBM as CCSID 13488. This
| CCSID (13488) has been used internally by DB2 UDB for storing graphic string
| data in euc-Japan and euc-Taiwan databases. CCSID 13488 and code page 1200
| both refer to UCS-2, and are handled the same way except for the value of
| their "double-byte" (DBCS) space:
 
|        CP/CCSID        Single Byte (SBCS) space      Double Byte (DBCS) space
|       ---------        ------------------------      ------------------------
|         1200                   N/A                           U+0020
|         13488                  N/A                           U+3000
 
|       NOTE: In a UCS-2 database, U+3000 has no special meaning.
 
| Regarding the conversion tables, since code page 1200 is a superset of CCSID
| 13488, the exact same (superset) tables are used for both.
 
| In IBM, UTF-8 has been registered as CCSID 1208 with growing character set
| (sometimes also referred to as code page 1208). As new characters are added
| to the standard, this number (1208) will not change either. 1208 is used as
| the multi-byte code page number for DB2's UCS-2/UTF-8 support.
 
| DB2 UDB supports UCS-2 as a new multi-byte code page. The MBCS code page
| number is 1208, which is the database code page number, and the code page of
| character string data within the database. The double-byte code page number
| (for UCS-2) is 1200 which is the code page of graphic string data within the
| database. When a database is created in UCS-2/UTF-8, CHAR, VARCHAR, LONG
| VARCHAR, and CLOB data, are stored in UTF-8, and GRAPHIC, VARGRAPHIC, LONG
| VARGRAPHIC, and DBCLOB data, are stored in UCS-2. We will simply refer to
| this as a UCS-2 database.
 
 
| 2.15.2.2  Creating a UCS-2 Database
 
| By default, databases are created in the code page of the application
| creating them. Therefore, if you create your database from a UTF-8 client
| (for example, the UNIVERSAL locale of AIX), or if DB2CODEPAGE environment
| variable on the client is set to 1208, your database will be created as a
| UCS-2 database. Alternatively, you can explicitly specify "UTF-8" as the
| CODESET name, and use any valid two letter TERRITORY code supported by DB2
| UDB. Refer to the Administration Guide, and the appendix on "NLS Support" for
| a complete list of territory codes supported by DB2 UDB.
 
| For example, to create a UCS-2 database from the CLP, with the territory code
| for United States, issue:
 
|     DB2 CREATE DATABASE dbname USING CODESET UTF-8 TERRITORY US
 
| The default collation sequence for a UCS-2 database is IDENTITY, which
| provides UCS-2 code point order. Therefore, by default, all UCS-2/UTF-8
| characters are ordered and compared according to their UCS-2 code point
| sequence.
 
| All cultural-sensitive parameters such as date/time format, decimal
| separator, and others, are based on the current territory of the client.
 
| A UCS-2 database allows connection from every single-byte and multi-byte code
| page supported by DB2 UDB. Code page character conversions between client's
| code page and UTF-8 are automatically performed by the database manager. Data
| in graphic string types, is always in UCS-2 and does not go through code page
| conversions. The Command Line Processor (CLP) environment is an exception. If
| you SELECT graphic string (UCS-2) data from the CLP, the returned graphic
| string data is converted (by the CLP) from UCS-2 to the code page of your
| client environment.
 
| Every client is limited by the character repertoire, the input methods, and
| the fonts supported by its environment, but the UCS-2 database itself accepts
| and stores all UCS-2 characters. Therefore, every client usually works with a
| subset of UCS-2 characters, but the database manager allows the entire
| repertoire of UCS-2 characters.
 
| When characters are converted from a local code page to UTF-8, there is a
| possibility of expansion in the number of bytes. There is no expansion for
| ASCII characters, but other UCS-2 characters expand by a factor of two or
| three. The number of bytes of each UCS-2 character in UTF-8 format can be
| determined from the above table (section about UTF-8).
 
 
| 2.15.2.3  Data Types
 
| All data types supported by DB2 UDB, are also supported in a UCS-2 database.
| In particular, graphic string data, is supported for UCS-2 database and is
| stored in UCS-2/Unicode. Every client, including SBCS clients, can work with
| graphic string data types in UCS-2/Unicode when connected to a UCS-2
| database.
 
| A UCS-2 database is like any MBCS database where character string data is
| measured in number of bytes. When working with character string data in
| UTF-8, one should not assume that each character is one byte. In multi-byte
| UTF-8 encoding, each ASCII character is one byte, but non-ASCII characters
| take two or three bytes each. This should be taken into account when defining
| CHAR fields. Depending on the ratio of ASCII to non-ASCII characters, a CHAR
| field of size n bytes, can contain anywhere from n/3 to n characters.
 
| Using character string UTF-8 encoding versus graphic string UCS-2 data type
| also has an impact on the total storage requirements. For a situation where
| the majority of characters are ASCII, with some non-ASCII characters in
| between, storing UTF-8 data may be a better alternative because the storage
| requirements are closer to one byte per character. On the other hand, for
| situations where the majority of characters are non-ASCII characters that
| expand to three-byte UTF-8 sequences (for example, ideographic characters),
| the UCS-2 graphic-string format may be a better alternative because every
| UCS-2 character requires exactly two bytes, rather than three bytes for each
| corresponding character in UTF-8 format.
 
| SQL scalar functions that operate on character strings, such as LENGTH,
| SUBSTR, POSSTR, MAX, MIN, and the like, in MBCS environments operate on
| number of "bytes" rather than number of "characters." The behaviour is the
| same in a UCS-2 database. For example, the LENGTH of a character string
| field, is the length in the number of bytes not the number of characters. For
| a description of the behaviour of these functions, refer to the DB2 UDB SQL
| Reference.
 
| SQL CHAR data types are supported by C language's "char" data type in user
| programs. SQL GRAPHIC data types are supported by "sqldbchar" in user C
| programs. Note that, for a UCS-2 database, "sqldbchar" data is always in
| big-endian (high byte first) format.  When an application program is
| connected to a UCS-2 database, character string data is converted between the
| application code page and UTF-8 by DB2 UDB, but graphic string data is always
| in UCS-2.
 
 
| 2.15.2.4  Identifiers
 
| In a UCS-2 database, all identifiers are in multi-byte UTF-8.  Therefore, it
| is possible to use any UCS-2 character in identifiers where the use of a
| character in the extended character set (for example, an accented character,
| or a multi-byte character) is allowed by DB2 UDB. Please refer to the
| appendix "Naming Rules" in the Administration Guide for details of which
| identifiers allow use of extended characters.
 
| Clients can enter any character which is supported by their SBCS/MBCS
| environment, and all the characters in the identifiers will be converted to
| UTF-8 by the database manager. Two points need to be taken into account when
| specifying National Language characters in identifiers in a UCS-2 database:
 
| 1.  Each non-ASCII character takes two or three bytes. Therefore, an n-byte
|     identifier, can only hold somewhere between n/3 and n characters,
|     depending the ratio of ASCII to non-ASCII characters. If you have only
|     one or two non-ASCII (for example, accented) characters, the limit is
|     closer to n characters, while for an identifier which is completely
|     non-ASCII (for example, in Japanese), only n/3 characters can be used.
 
| 2.  If identifiers are to be entered from different client environments, they
|     should be defined using the common subset of characters available to
|     those clients. For example, if a UCS-2 database is to be accessed from
|     Latin-1, Arabic, and Japanese environments, all identifiers should
|     realistically be limited to ASCII.
 
 
| 2.15.2.5  UCS-2 Literals
 
| UCS-2 literals can be specified in two ways:
 
| 1.  As a GRAPHIC string constant using the G'...', or N'....' format as
|     described in the SQL Reference, Chapter 3 "Language Elements," the
|     section  "Constants," the subsection "Graphic String Constants." Any
|     literal specified in this way will be converted by the database manager
|     from the application code page to UCS-2.
 
| 2.  As a UCS-2 hexadecimal string, using the UX'....' or GX'....' format. The
|     constant specified between quotes after UX or GX must be a multiple of 4
|     hexadecimal digits. Each four digits represent one UCS-2 code point.
 
| When using the Command Line Processor (CLP), the first method is easier if
| the UCS-2 character exists in the local application code page (for example,
| for entering any code page 850 character from a terminal that is using code
| page 850). The second method should be used for characters which are outside
| the application code page repertoire (for example, for specifying Japanese
| characters from a terminal that is using code page 850).
 
 
| 2.15.2.6  Pattern Matching in a UCS-2 Database
 
| Pattern matching is one area where the behaviour of existing MBCS databases
| is slightly different from the behaviour of a UCS-2 database.
 
| For MBCS databases in DB2 UDB, the current behaviour is as follows: If the
| match-expression contains MBCS data, the pattern can include both SBCS and
| MBCS characters. The special characters in the pattern are interpreted as
| follows:
 
| o   An SBCS underscore refers to one SBCS character.
 
| o   A DBCS underscore refers to one MBCS character.
 
| o   A percent (either SBCS or DBCS) refers to a string of zero or more SBCS
|     or MBCS characters.
 
| If the match-expression contains graphic string DBCS data, the expressions
| contain only DBCS characters. The special characters in the pattern are
| interpreted as follows:
 
| o   A DBCS underscore refers to one DBCS character.
 
| o   A DBCS percent refers to a string of zero or more DBCS characters.
 
| In a UCS-2 database, there is really no distinction between "single-byte" and
| "double-byte" characters; every UCS-2 character occupies two bytes. Although
| the UTF-8 format is a "mixed-byte" encoding of UCS-2 characters, there is no
| real distinction between SBCS and MBCS characters in UTF-8. Every character
| is a UCS-2 character, irrespective of its number of bytes in UTF-8 format.
| When specifying a character string, or a graphic string expression, an
| underscore refers to one UCS-2 characters and a percent refers to a string of
| zero or more UCS-2 characters.
 
| On the client side, the character string expressions are in the code page of
| the client, and will be converted to UTF-8 by the database manager. SBCS
| client code pages, do not have any DBCS percent or DBCS underscore, but every
| supported code page contains a single-byte percent (corresponding to U+0025)
| and a single-byte underscore (corresponding to U+005F). The interpretation of
| special characters for a UCS-2 database is as follows:
 
| o   An SBCS underscore (corresponding to U+0025) refers to one UCS-2
|     character in a graphic string expression, or to one UTF-8 character in a
|     character string expression.
 
| o   An SBCS percent (corresponding to U+005F) refers to a string of zero or
|     more UCS-2 characters in a graphic string expression, or to a string of
|     zero or more UTF-8 characters in a character string expression.
 
| DBCS code pages, additionally support a DBCS percent sign (corresponding to
| U+FF05) and a DBCS underscore (corresponding to U+FF3F). These characters
| have no special meaning for a UCS-2 database.
 
| For the optional "escape-expression" which specifies a character to be used
| to modify the special meaning of the underscore and percent characters, only
| ASCII characters, or characters that expand into a two-byte UTF-8 sequence,
| are supported. If you specify an ESCAPE character which expands to a
| three-byte UTF-8 value, you will get an error message (SQL0130N error,
| SQLSTATE 22019).
 
 
| 2.15.2.7  IMPORT/EXPORT/LOAD Considerations
 
| The DEL, ASC, and PC/IXF file formats are supported for a UCS-2 database as
| described in this section. The WSF format is not supported. Also, autoloader
| is not supported for a UCS-2 database at the present time.
 
| When exporting from a UCS-2 database to an ASCII delimited (DEL) file, all
| character data is converted to the application code page. Both character
| string and GRAPHIC string data are converted to the same SBCS or MBCS code
| page of the client. This is the existing behaviour for the export of any
| database, and cannot be changed because the entire ASCII delimited file can
| have only one code page. Therefore, if you export to an ASCII delimited file,
| only those UCS-2 characters that exist in your application code page would be
| saved. Other characters are replaced with the default substitution character
| for the application code page. For UTF-8 clients (code page 1208) there is no
| data loss because all UCS-2 characters are supported by UTF-8 clients.
 
| When importing from an ASCII file (DEL or ASC) to a UCS-2 database, character
| string data is converted from the application code page to UTF-8, and GRAPHIC
| string data is converted from the application code page to UCS-2.  There is
| no data loss. If you want to import ASCII data that has been saved under a
| different code page, you should switch the data file code page before issuing
| the import command. One way to accomplish this, is to set DB2CODEPAGE to the
| code page of the ASCII data file.
 
| The range of valid ASCII delimiters, for SBCS and MBCS clients is identical
| to what is currently supported by DB2 UDB for these clients. The range of
| valid delimiters for UTF-8 clients is 0x01 to 0x7F, with the usual
| restrictions. Refer to the "IMPORT/EXPORT/LOAD Utility File Formats" appendix
| in the Command Reference for a complete list of these restrictions.
 
| When exporting from a UCS-2 database to a PC/IXF file, character string data
| is converted to the SBCS/MBCS code page of the client. GRAPHIC string data is
| not converted and is stored in UCS-2 (code page 1200). There is no data loss.
 
| When importing from an PC/IXF file to a UCS-2 database, character string data
| is assumed to be in the SBCS/MBCS code page stored in the PC/IXF header and
| GRAPHIC string data is assumed to be in the DBCS code page stored in the
| PC/IXF header. Character string data is converted by the IMPORT utility from
| the code page specified in the PC/IXF header to the code page of the client,
| and then from the client code page to UTF-8 (by the INSERT statement).
| GRAPHIC string data is converted by the IMPORT utility from the DBCS code
| page specified in the PC/IXF header directly to UCS-2 (code page 1200).
 
| LOAD directly places the data into the database and by default, assumes data
| in ASC or DEL files is in the code page of the database. Therefore, by
| default no code page conversion takes place for ASCII files. When the code
| page of the data file has been explicitly specified (using the MODIFIED BY
| codepage=x command parameter), LOAD uses this information to convert from the
| specified code page x into the database code page before loading the data.
| For PC/IXF files, LOAD always converts from the code pages specified in the
| IXF header to the database code page (1208 for CHAR, 1200 for GRAPHIC).
 
| The code page of DBCLOB files (as specified using the MODIFIED BY lobsinfile
| command parameter) is always 1200 for UCS-2. The code page of the CLOB files
| is the same as the code page of the data files being imported, loaded or
| exported. For example, for load or import using PC/IXF format, the CLOB file
| is assumed to be in the code page specified by the PC/IXF header. If the
| DBCLOB file is in ASC or DEL format, for LOAD the CLOB data is assumed to be
| in the code page of the database (unless explicitly specified otherwise using
| the MODIFIED BY codepage=x command parameter), and for IMPORT it is assumed
| to be in the client application code page.
 
| The NOCHECKLENGTHS option is always set to TRUE for a UCS-2 database because
| in a UCS-2 database, any SBCS can be connected to the database for which
| there is no DBCS code pages; and also because character strings in UTF-8
| format usually have different lengths than corresponding lengths in client
| code pages.
 
 
| 2.15.2.8  Incompatibilities
 
| For an application connected to a UCS-2 database, the graphic string data is
| always in UCS-2 (code page 1200). For applications connected to non UCS-2
| databases, the graphic string data is in the applications DBCS code page; or,
| not allowed if the application code page is SBCS. For example, when a 932
| client is connected to a Japanese non UCS-2 database, then the graphic string
| data is in code page 301. But for the 932 client applications connected to a
| UCS-2 database, the graphic string data is in UCS-2.
 
 
| 2.16  USING VIRTUAL INTERFACE (VI) ARCHITECTURE
 
| DB2 UDB now supports two implementations of the VI architecture: One from
| GigaNet and the other from ServerNet. Each is presented in this section.
 
 
| 2.16.1  OVERVIEW OF VIRTUAL INTERFACE ARCHITECTURE
 
| Virtual Interface (VI) Architecture is the inter-node communication protocol
| alternative to TCP/IP in a Windows NT massively parallel processing (MPP)
| environment. VI is a new communication architecture that was developed
| jointly by Intel, Microsoft, and Compaq to improve performance over a System
| Area Network (SAN). For more information on the architecture, visit:
 
|      http://www.viarch.org
 
| There are some similarities between the Public Interconnect which uses as an
| example Ethernet and TCP/IP and the Private Interconnect which uses a Network
| Interface Card and a protocol. The Network Interface Card and protocol used
| in this instance is a GigaNet Network Interface Card and the VI protocol.
 
| VI Architecture has low latency and high bandwidth. In a
| communication-intensive environment, using VI Architecture improves the
| overall system throughput. The greater the number of nodes in the cluster,
| and the greater the amount of data transfered, the greater the benefit from
| using VI Architecture.
 
| DB2 UDB supports VI Architecture implementations that comply with the Virtual
| Interface Architecture Specification, Version 1.0; the Intel Virtual
| Interface (VI) Architecture Developers' Guide, Version 1.0; and pass the
| "Virtual Interface Architecture Conformance Suite." The specification is
| found at:
 
|      http://www.intel.com/design/servers/vi/the_spec/specification.htm
 
| The Developer's Guide and information on the conformance suite is found at:
 
|      http://www.intel.com/design/servers/vi/developer/ia_imp_guide.htm
 
 
| 2.16.2  INSTALL DB2 UNIVERSAL DATABASE VERSION 5.2 (EEE)
 
| Detailed installation information is found in DB2 Enterprise - Extended
| Edition for Windows NT Quick Beginnings.
 
| If you are not sure of the service level of the DB2 UDB product you have
| already installed, you can determine the UDB service level by going into the
| registry:
 
| 1.  Enter regedt32 from a command prompt to edit the registry
 
| 2.  Choose HKEY_LOCAL_MACHINE
 
| 3.  Choose SOFTWARE
 
| 4.  Choose IBM
 
| 5.  Choose DB2
 
| 6.  Choose DB2 UDB Enterprise - Extended Edition
 
| 7.  Choose CurrentVersion
 
| 8.  Record the Service Level
 
| If you contact DB2 Service and Support about VI, the service level will be
| helpful to determine your installed DB2 level of code including any FixPaks.
 
| This product must be installed in each of the partitions/nodes using the
| Virtual Interface Protocol. During the installation procedure, when prompted
| choose "This machine will be an instance owning node" on each of the
| partitions/nodes.
 
| Update the hosts file with the IP address and host name for each of the
| partitions/nodes. The hosts file is found under the
| "\winnt\system32\drivers\etc\" directory on the drive where the operating
| system was installed. The hosts file must be updated on each of the nodes.
 
| Create the partitioned database (MPP) instance using the instance create
| utility. Choose one machine to act as the co-ordinator node. On this machine,
| open a DB2 Command Window and enter:
 
|      db2icrt <instance_name> /mpp /u:<username>,<password>
 
| This machine is then known as the co-ordinator node or the instance-owning
| machine. Node 0 is automatically created on this machine.
 
| On the other partitions/nodes in the database, open a DB2 Command Window and
| enter:
 
|      db2ncrt /n:<node_number> /u:<username>,<password>
|                 /i:<instance_name> /o:<instance_owner_name>
 
| The node_number is used to uniquely identify the database partition server
| within the database environment. The number must be from 1 to 999. The
| instance_owner_name is the computer name of the instance-owning machine
| (co-ordinator node).
 
| Testing the installation and create an index:
 
| 1.  Open a DB2 Command Window.
 
| 2.  Enter "set DB2INSTANCE=<instance_name>"
 
| 3.  Ensure the database manager starts on all nodes by entering: "db2start"
 
| 4.  Create a sample database by entering: "db2sampl"
 
| 5.  Connect to the sample database by entering: "db2 connect to sample"
 
| 6.  Try a few SELECT statements with the sample database.
 
| When problems occur in this environment, you can take action based on the
| type of problem as presented below:
 
| o   Instance creation fails.
 
|     Ensure "c:\profiles" is present and is present with each of the share
|     name "profiles." Ensure all partitions are "pingable" from the
|     co-ordinator node.
 
| o   DB2START fails.
 
|     Review the explanation for the returned error code by using the "db2 ?
|     sqlxxxx" command. There will be a suggested action associated with this
|     error which you should follow.
 
|     A system error may be returned. If this is the case, use a "db2stop" and
|     retry the "db2start". If the problem persists, attempt to reboot on all
|     partitions and then retry.
 
|     Ensure all partitions have the same date, time, and time zone. The time
|     does not need to be identical: within one hour is sufficient.
 
|     Ensure all the partitions are in one domain and that the user name and
|     password used belong to the following groups:
 
|         -   On the domain controller:
 
|         -   Administrators
 
|         -   Domain Administrators
 
|         -   Domain Users
 
|         -   Users
 
|         -   On other machines:
 
|         -   Administrators
 
|         -   Users
 
|     Review the contents of the "Control panel-->Services" to ensure that all
|     the "DB2:<instance_name> -X" services have the correct DB2ADMIN account
|     information.
 
| o   Command line variable has not been initialized:
 
|     Ensure you are running the command in a "DB2 command window." The title
|     of this window is "DB2 CLP."
 
| o   The "rah" command returns immediately without executing the commands
|     specified:
 
|     Run "db2set -g DB2TEMPDIR=C:\TMP" on all machines in the instance. Ensure
|     the DB2 Remote Command Service is started and with the correct DB2ADMIN
|     account information. Finally, ensure "c:\temp" and "c:\tmp" are present.
 
 
| 2.16.3  RUNNING DB2 UDB V5.2 FOR WINDOWS NT WITH GIGANET INTERCONNECT
 
| IBM announced support for Virtual Interface (VI) Architecture with DB2 UDB
| EEE V5.2. The documentation provided no specifics regarding the VI products
| that have been tested with DB2. This document provides specific information
| for GigaNet Interconnect.
 
| To find out about other products adhering to VI Architecture and supported by
| DB2 UDB EEE V5.2, please contact the DB2 UDB support organization at
| http://www.software.ibm.com/data or call 1-800-237-5511 (only in the U.S.A).
 
| To find out about GigaNet products, or to contact GigaNet Service and
| Support, please use the following URL: http://www.giganet.com/
 
 
| 2.16.3.1  Setup Procedure for GigaNet Interconnect
 
| The list of the hardware and software required to setup this environment
| include the following products:
 
| o   GigaNet GNN1000 Network Interface Card
 
| o   GigaNet GNX5000 Switch
 
| o   GigaNet GNCxx11 Copper Interconnect Cables
 
| o   GigaNet cLAN Software, Version 2.0.
 
| The steps required to ensure that GigaNet Interconnect can work with DB2 UDB
| are shown below. Each step is a summary of what is required at each step: all
| of the details associated with each step are not presented here. You should
| also use the referenced documenation at each step which does provide detailed
| instructions and direction needed.
 
| Each GigaNet GNN1000 is packaged with a GigaNet cLAN Software CD-ROM. The
| CD-ROM contains all of the necessary software to set-up the GigaNet
| Interconnect. In addition, the CD-ROM also contains the VI Architecture SDK
| and Adobe Acrobat Reader. These two items are only needed by those
| individuals that are developing VI-enabled applications.
 
| Summary of steps:
 
| 1.  Install Adapter Cards
 
| 2.  Install Switches and Cables
 
| 3.  Install Adapter Drivers
 
| 4.  Install cLAN Management Console
 
| 5.  Test the Interconnect
 
| Here are the steps:
 
| 1.  Install the GigaNet GNN1000 Network Interface Card. Please refer to the
|     GigaNet GNN1000 User Guide for installation instructions.
 
| 2.  Install the GigaNet GNX5000 Switch and Cables. Please refer to the
|     GigaNet GNX5000 User Guide for installation instructions.
 
| 3.  Install the GigaNet GNN1000 Adapter Driver software on each node
|     connected to the GNX5000 Switch. Please refer to the GigaNet GNN1000 User
|     Guide for installation instructions. Here are additional details if you
|     are installing drivers provided by GigaNet:
 
|     a.  Remove any previous version of the GNN1000 Driver already installed.
|         Removal requires the node to be re-booted.
 
|     b.  Use "Start>Setting>Control Panel>Networks>Adapters>Add" to install
|         the driver.
 
|     c.  Click HAVE DISK... and specify the Driver directory on the CD-ROM.
|         For example, if F: is your CD-ROM drive, then you would use
|         "F:\Driver"
 
|     d.  Select "GNN1000 NDIS Adapter" and then click OK.
 
|     e.  Configure Network protocols to complete the installation.
 
|     GigaNet Adapter Driver software is also available on GigaNet's web site,
|     http://www.giganet.com. Please refer to the download and installation
|     instructions found on the support page of GigaNet's web site.
 
|     The installation of the GNN1000 Adapter Driver causes the node to
|     re-boot.
 
| 4.  The GigaNet cLAN Management Console (GMC) can be used to test the
|     integrity of the GigaNet Interconnect. The GigaNet cLAN Management
|     Console is comprised of two parts: the Console, and the Agent. The Agent
|     must be installed on all nodes in the cluster. The Console can be
|     installed on any network node that has access to the nodes in the
|     cluster. The most versatile and recommended installation is that which
|     has both the Console and the Agent installed on each node in the cluster.
 
|     Install the GigaNet cLAN Management Console. Please refer to the GigaNet
|     GNN1000 User Guide for installation instructions and additional
|     information about the cLAN Management Console. Here are additional
|     details on the installation procedure:
 
|     a.  Insert the cLAN Software CD into the CD-ROM drive.
 
|     b.  Wait for the CD automatic installation menu to appear.
 
|     c.  Click on "Install cLAN Management Console."
 
|     d.  Repeat this installation procedure on each remaining node in the
|         cluster.
 
|     GigaNet cLAN Management Console software is also available on GigaNet's
|     web site, http://www.giganet.com. Please refer to the download and
|     installation instructions found on the support page of GigaNet's web
|     site.
 
|     The installation of the cLAN Management Console may cause the node to
|     re-boot.
 
| 5.  Test that the GigaNet Hardware is working. This can be done by doing the
|     following:
 
|     a.  Open the GMC. ("Programs>GigaNet>cLAN Management Console")
 
|     b.  A dialog box is displayed showing all accessible machines in the LAN.
|         Press ESC.
 
|     c.  Select CONSOLE>LOCAL from the menu bar.
 
|     d.  Confirm that all the members in the cluser are shown and that they
|         are all "Active."
 
|     e.  Select UTILITIES>VI THROUGHPUT from the menu bar. This will run a
|         throughput test to check that the data is actually going through the
|         hardware.
 
|     f.  Enter in uppercase letters the computer names of the two nodes you
|         wish to use in the test. Identify the local node as the source node.
 
|     g.  Click START MEASURING. You should see data being transferred at a
|         rate of at least 65 MB per second.
 
|     h.  Click STOP MEASURING to stop the connection test.
 
|     i.  Repeat the test for the other nodes in the cluster by measuring
|         throughput between the local node (Source) and the other nodes
|         (Sink).
 
|     If the connection test does not appear to be working, refer to the
|     troubleshooting sections of the GigaNet GNN1000 User Guide and the
|     GigaNet GNX5000 User Guide.
 
 
| 2.16.3.2  Implement DB2 to Run Using VI
 
| On each database partition server in the instance, set the following DB2
| registry variables and carry out the following tasks:
 
| o   Set DB2_VI_ENABLE=ON
 
| o   Set DB2_VI_DEVICE=nic0
 
| o   Set DB2_VI_VIPL=vipl.dll
 
| o   Enter "db2start" on the MPP instance.
 
| o   Review the db2diag.log file. There should be one message for each
|     partition stating that "VI is enabled."
 
 
| 2.16.4  RUNNING DB2 UDB V5.2 FOR WINDOWS NT WITH SERVERNET INTERCONNECT
 
| IBM announced support for Virtual Interface (VI) Architecture with DB2 EEE
| UDB V5.2. The documentation provided no specifics regarding the VI products
| that have been tested with DB2. This document provides specific information
| for ServerNet Interconnect.
 
| To find out about other products adhering to VI Architecture and supported by
| DB2 UDB EEE V5.2, please contact the DB2 UDB support organization at
| http://www.software.ibm.com/data or call 1-800-237-5511 (only in the U.S.A).
 
| To find out about ServerNet products, or to contact ServerNet Service and
| Support, please use the following URL: http://www.servernet.com/
 
 
| 2.16.4.1  Setup Procedure for ServerNet Interconnect
 
| The list of the hardware and software required to setup this environment
| include the following products:
 
| o   ServerNet PCI Adapter Driver (SPAD), (product ID T0089), version 1.3.5 or
|     later
 
| o   ServerNet Switch 1
 
| o   ServerNet Area Network Manager (SANMan), (product ID T0087), version
|     1.1.3 or later.
 
| The following are the steps required to ensure that ServerNet Interconnect
| can work with DB2 UDB. Each step is a summary of what is required at each
| step: all of the details associated with each step are not presented here.
| You should also use the referenced documenation at each step which does
| provide detailed instructions and direction needed.
 
| The steps shown below also assume that you are only using up to six (6) nodes
| in the cluster. Contact ServerNet if you have a requirement to use more than
| six nodes.
 
| Here are the steps:
 
| 1.  Install the ServerNet Network Interface Card. Please refer to the
|     ServerNet-I Virtual Interface Software Release Document, (product ID
|     N0031) for installation instructions.
 
| 2.  Install the ServerNet Switch 1. Please refer to the ServerNet-I Virtual
|     Interface Software Release Document, (product ID N0031) for installation
|     instructions.
 
| 3.  Uninstall previous ServerNet drivers. (Skip this step if this is your
|     first time installing ServerNet.)
 
|     a.  Open the Network control panel. ("Start>Setting>Control
|         Panel>Network")
 
|     b.  Click on the ADAPTERS TAB.
 
|     c.  Remove Tandem ServerNet PCI Adapter Driver.
 
|     d.  Click on the SERVICES TAB.
 
|     e.  Remove SANMan.
 
|     f.  Click on the PROTOCOLS TAB.
 
|     g.  Remove Tandem ServerNet-I VI Protocol.
 
| 4.  Install the Tandem ServerNet PCI Adapter Driver. Here are additional
|     details if you are installing using the software CD provided by
|     ServerNet:
 
|     a.  Open the Network control panel. ("Start>Setting>Control
|         Panel>Network")
 
|     b.  Click on the ADAPTERS TAB. (The Adapters screen appears.)
 
|     c.  Ensure the new ServerNet driver is placed in a separate drive and/or
|         directory. Then, from the command prompt referencing the correct
|         drive and/or directory, type "ernnn.exe -d" to start the
|         self-extracting program. ("ernnn.exe" is the name of the Engineering
|         Release followed by a number -- ERnnn.EXE -- that identifies the
|         specific version of the ServerNet driver to be installed.)
 
|     d.  Change to the drive and/or directory where the extracted files are
|         located. Change to the "Spad n.n.n \ Free" sub-directory (where
|         "n.n.n" is the specific version of the product). (If you are working
|         in a troubleshooting or a development environment, then change to the
|         "Spad n.n.n \ Checked" sub-directory instead of the "Spad n.n.n \
|         Free" sub-directory.)
 
|     e.  Rename the "oemsetup.multi_node" file to "oemsetup.inf."
 
|     f.  Choose ADD in the Adapters Tab. (The Select Adapters screen appears.)
 
|     g.  Click HAVE DISK.... (The Insert Disk screen appears.)
 
|     h.  Enter the drive and/or directory where the oemsetup.inf file is
|         located.
 
|     i.  Ensure the dialog box shows "Tandem ServerNet PCI Adapter Driver" and
|         then click OK. Ensure the list of adapters now shows the ServerNet
|         adapter. Click CLOSE.
 
|     j.  Choose YES to restart the computer. Or, select NO and continue
|         installing SANMan and the VI Software Developer's Kit (SDK).
 
| 5.  Install SANMan. Here are additional details if you are installing using
|     the software CD provided by ServerNet:
 
|     a.  Open the Network control panel. ("Start>Setting>Control
|         Panel>Network")
 
|     b.  Click on the SERVICES TAB. (The Services screen appears.)
 
|     c.  Ensure the new ServerNet driver is placed in a separate drive and/or
|         directory. Then, from the command prompt referencing the correct
|         drive and/or directory, type "ernnn.exe -d" to start the
|         self-extracting program. ("ernnn.exe" is the name of the Engineering
|         Release followed by a number -- ERnnn.EXE -- that identifies the
|         specific version of the ServerNet driver to be installed.)
 
|     d.  Choose ADD in the Services Tab. (The Select Services screen appears.)
 
|     e.  Change to the drive and/or directory where the extracted files are
|         located. Change to the "SANMan n.n.n \Free" sub-directory (where
|         "n.n.n" is the specific version of the product). (If you are working
|         in a troubleshooting or a development environment, then change to the
|         "SANMan n.n.n \ Checked" sub-directory instead of the "SANMan n.n.n \
|         Free" sub-directory.)
 
|     f.  Determine if the Switch is X or Y by looking at the light on the
|         Switch.  One light says "X," and the one light says  "Y."
 
|     g.  If an X Switch, select X=1 and Y=0. Ensure all cables are connected
|         to the X port on the network cards.
 
|     h.  If a Y Switch, select X=0 and Y=1. Ensure all cables are connected to
|         the Y port on the network cards.
 
|     i.  Provide the port number of the switch to which the network card on
|         the current machine is connected.
 
|     j.  Select "PC" for all six (6) ports.
 
| 6.  Install the Virtual Interface Protocol. Here are additional details if
|     you are installing using the software CD provided by ServerNet:
 
|     a.  Open the Network control panel. ("Start>Setting>Control
|         Panel>Network")
 
|     b.  Click on the PROTOCOLS TAB. (The Network Protocols screen appears.)
 
|     c.  Ensure the new ServerNet driver is placed in a separate drive and/or
|         directory. Then, from the command prompt referencing the correct
|         drive and/or directory, type "ernnn.exe -d" to start the
|         self-extracting program. ("ernnn.exe" is the name of the Engineering
|         Release followed by a number -- ERnnn.EXE -- that identifies the
|         specific version of the ServerNet driver to be installed.)
 
|     d.  Choose ADD in the Protocols Tab. (The Select Network Protocols screen
|         appears.)
 
|     e.  Click HAVE DISK.... (The Insert Disk screen appears.)
 
|     f.  Enter the drive and/or directory where the extracted files are
|         located.
 
| 7.  Test that the ServerNet Hardware is working. There are no test programs
|     available. Instead, simply use DB2 to test the ServerNet hardware.
 
|     If the hardware does not appear to be working, refer to the ServerNet-I
|     Virtual Interface Software Release Document, (product ID N0031) for
|     additional troubleshooting help.
 
 
| 2.16.4.2  Implement DB2 to Run Using VI
 
| On each database partition server in the instance, set the following DB2
| registry variables and carry out the following tasks:
 
| o   Set DB2_VI_ENABLE=ON
 
| o   Set DB2_VI_DEVICE=nic0
 
| o   Set DB2_VI_VIPL=vipl.dll
 
| o   Enter "db2start" on the MPP instance.
 
| o   Review the db2diag.log file. There should be one message for each
|     partition stating that "VI is enabled."
 
 
  3.0  API REFERENCE (ŧ)
 
 
  3.1  SQLECREA - CREATE DATABASE
 
  To specify the IDENTITY collating sequence when creating a database, specify
  SQL_CS_NONE (which implements a binary collating sequence) in the database
  description block (SQLEDBDESC) structure.
 
 
  3.2  SQLUBKP - BACKUP DATABASE
 
  In the description of the pMediaTargetList parameter (value SQLU_ADSM_MEDIA),
  the sentence "No additional input is required." should be replaced with:
 
       If an SQLU_MEDIA_ENTRY structure is not being used to specify
       a path for the backup image, initialize the MEDIA pointer in
       the SQLU_MEDIA_LIST_TARGETS structure to NULL.
 
 
  3.3  SQLUGTPI - GET TABLE PARTITIONING INFORMATION
 
  The following section is missing from the API description:
 
       Required Connection
          Database
 
 
  3.4  SQLUHGET - RETRIEVE DDL INFORMATION FROM THE HISTORY FILE (NEW API)
 
  There is no callerac parameter available in the sqluhgne API with
  which to retrieve DDL information recorded in the history file. To retrieve this information, a new API has been created:
 
       SQL_API_RC SQL_API_FN
         sqluhget(
           unsigned short Handle,            // IN: Handle returned from sqluhops
           unsigned short CallerAction,      // IN: For special handling of command field
           unsigned long Reserved_1,         // Reserved for future use
           struct sqluhinfo * pHistoryInfo,  // IN/OUT: Pointer to entry data
           struct sqluhadm * pAdminInfo,     // IN/OUT: Pointer to enhanced adminstrative data
           void * pReserved,                 // Reserved for future use
           struct sqlca * pSqlca);           // SQLCA
 
  The parameters of this API have the same meanings as the corresponding
  parameters of the sqluhgne API, with the exception of CallerAction, which is new. The reserved fields of this API, Reserved_1 and pReserved, must be initialized
  to 0 and NULL, respectively. The possible values of CallerAction are:
 
       SQLUH_GET_NEXT_ENTRY
          Retrieves the next matching entry.
          No DDL information is returned to the caller, but the length
          field of the sqluhadm.command structure will be set to the length
          of any DDL field that exists for that entry.
       SQLUH_GET_DDL
          If this action is passed to the API immediately after
          fetching an entry, the DDL data associated with that entry is returned.
          It is the responsibility of the caller to indicate - through
          the sqluhadm.command.length field - the number of bytes
          of memory which have been allocated for DDL data to be
          returned through the sqluhadm.command.data field.
       SQLUH_GET_NEXT_ENTRY_DDL
          Retrieves the next matching entry.
          It also returns any DDL information which has been recorded for the entry.
          The caller must indicate - through the sqluhadm.command.length field -
          the number of bytes of memory which have been allocated for
          the sqluhadm.command.data field.
 
 
  3.5  SQLUHGNE - GET NEXT RECOVERY HISTORY FILE ENTRY
 
  The second parameter of this API, which was "int callerac", is now
  "struct sqluhadm *pAdminInfo". The API definition is:
 
       SQL_API_RC SQL_API_FN
         sqluhgne(
           unsigned short Handle,
           struct sqluhadm * pAdminInfo,
           struct sqluhinfo * pHistoryInfo,
           struct sqlca * pSqlca);
 
 
  3.6  SQLUHOPS - OPEN RECOVERY HISTORY FILE SCAN
 
  The description of the SQLUH_LIST_ADM_ALTER_TABLESPACE caller action
  should be changed to:
 
       Select only the ALTER TABLESPACE records that pass the other
       filters.  The DDL field associated with an entry will not be
       returned.  To retrive the DDL information for an entry,
       sqluhgne must be called with a caller action of SQLUH_GET_DDL
       immediately after the entry is fetched.
 
 
  3.7  SQLULOAD - LOAD
 
  This API affects only the partition to which an application is directly
  connected; the load utility operates on a single database partition only.
 
  The Usage Notes section incorrectly states that "summary tables" that are
  dependent on the tables being loaded are placed in check pending state. It
  should state that "summary tables defined with REFRESH IMMEDIATE", that are
  dependent on the tables being loaded, are placed in check pending state.
 
 
  3.8  SQLE-CLIENT-INFO (CONNECTION SETTINGS)
 
  The valid entries for the SQLE-CLIENT-INFO TYPE element should be changed to:
 
       SQLE_CLIENT_INFO_USERID
       SQLE_CLIENT_INFO_WRKSTNNAME
       SQLE_CLIENT_INFO_APPLNAME
       SQLE_CLIENT_INFO_ACCTSTR
 
 
  3.9  SQLFUPD
 
  Table 46, "Updateable Database Manager Configuration Parameters" has an
  additional entry for a new parameter, catalog_noauth:
 
       Parameter Name   Token                     Token Value   Data Type
       --------------   -----                     -----------   ---------
       catalog_noauth   SQLF_KTN_CATALOG_NOAUTH   314           Uint16
 
 
  3.10  SQLUHINFO (SQLUHADM)
 
  The definition of the SQLUHADM structure should be changed to:
 
       SQL_STRUCTURE sqluhadm
       {
         char           end_time[SQLUH_TIMESTAMP_SZ+1]; // OUT: Completion time stamp of the event
         char           id[SQLUH_ID_SZ+1];              // OUT: Unique identifier of a dropped table
         struct sqlca   event_sqlca;                    // OUT: SQLCA associated with the entry
         struct sqlchar command;                        // IN/OUT: DDL command information
       };
 
  NOTES:
 
  1.  Currently, command data is recorded only for ALTER TABLESPACE events.
 
  2.  It is the responsibility of the caller to initialize the
      sqluhadm.command.length field to the number of bytes of memory which
      have been allocated for this field.
 
  3.  The macro SQLUHADMSIZE(n), defined in sqlutil, is provided to help
      determine how much memory is required for an sqluhadm structure
      with space for n characters of command data.
 
 
  3.11  SQLUVPUT - WRITING DATA TO DEVICE
 
  Change "struct Init_output *," in the "C API Syntax" section to "struct Data
  *,".
 
 
  4.0   WINDOWS ů OS/2 Ȣ
 
 
| 4.1  WINDOWS NT ĸ DB2START ȴع
 
|  "db2start"  ( "NET START" ) ٯ Windows NT
| ĸջܡȢ롢ΪǶ˷ٯЬ
| ǶˡϯġNT ǵ뢡 "DB2DIAG.LOG" 
|  "db2start" ̯Ƕ먡Զҡ
| Windows NT ĸջܡȢҳ NT Τ٭ȢΪ
| ɳٯΤ٭Ȣ㡢Ь먡
 
 
  4.2   WINDOWS NT  WINDOWS 95  SQL Ȣ (IBM
       VISUALAGE FOR COBOL)
 
  VisualAge COBOL for Windows NT  2.2  "cob2 -dll" ȴ
  "x:\sqllib\samples\cobol\bldvacbs.bat" ǡ DLL 
  Ķ
 
       cob2 -dll %1.obj db2api.lib
 
  Ⱥ VisualAge COBOL Ķ먡
 
       먡 "E:\sqllib\samples\cobol\db2api.lib"
 
  Ⱥ VisualAge COBOL for Windows NT ġԶ FixPak 
  ߡ
  紡 FixPak ƫЩȺƫĶ̽
  ĸ
 
       ilib  /nol /gi:%1 %1.obj
     ilink  /free /nol /dll  db2api.lib %1.exp %1.obj iwzrwin3.obj
 
 
  4.3   JAVA Ȣ APPLETS
 
 
  4.3.1   MICROSOFT SDK FOR JAVA  WINDOWS 
 
  ȴġԶƫ Windows ɢ˷ĸ DB2 SQLJ 
  ġ Java Development KitȺƵɢ DB2JVIEW
  ħ sqlj ȴ DB2 SQLJ  (db2profcdb2profp
  profdb  profp)Ϊ DB2JVIEW=1 Microsoft SDK for
  Javaľ멡ΪƵ "profp"ҳ "jview
  sqlj.runtime.profile.util.ProfilePrinter"Ϊ DB2JVIEW=0
   Sun JDKľ멡ΪƵ "profp"
  ҳ "java sqlj.rutnime.profile.util.ProfilePrinter"
  "sqlj" ܩȺġ Sun JDK 硤
 
 
| 4.3.2  ڷ׻ůΤȢ AND SQLJ Щ SCRIPT
 
| شħġ Java Ȣաʩ UDF Ȣ
| šȩԶɱڷ׻ΤȢƫڷ׻ĸϯڷ
| ̽Τĸʩ UDFЩ̯̿Ķ
 
| DB2STP         ܢΤȢƵڷ׻Ȣ DB2SPCLI
 
| DB2UDF         ܢΤȢƵڷ׻Ȣ DB2UDCLI
 
| STP            ܢΤȢƵڷ׻Ȣ STPCLI
 
| UDF            ܢڷ׻ȢƵΤȢ UDFSRV
 
| ǡ Script  "embprep" Щ SQLJ Ȣ
| "makefile" Ƶ Java  SQL Ȣĸ "db2profc"
| թջϯ ID ȴƫſȺɷ
| ڷ׻ڷ̯ DB2 Τĸջ
 
 
  5.0   UNIX Ȣ
 
 
  5.1  SCO UNIXWARE 7
 
 
| 5.1.1  ʩ̪
 
|  SCO UnixWare 7 㡢ڷ׻Ȣ̪
| ڷ׻ȢƵʩջҳħ̪ڷ׻
| ȢƶġԶ "-Wl,-Bexport"ƫ
| ȴڷ׻Ȣ̪Զ̯ SCO UnixWare
| 7 C ů C++  make 㡢ڷ׻Ȣ "fillcli" (
| Ƶʩջ "fillsrv" make ůȢ
|  "sqllib/samples/c" ů "sqllib/samples/cpp" 㡤
 
 
  5.1.2  MICRO FOCUS COBOL 
 
   "Micro Focus COBOL" 㡢"bldmfcc" Script 
   "-lthread" ѡҳջī DB2 ջ
  "-ldb2" 
 
 
  5.2   JAVA Ȣ APPLETS
 
 
  5.2.1  躡HP-UXSCO UNIXWARE 7
 
  ܢ HP-UX ĸ Java HP-UX Developer's Kit
  for Java Release 1.1.3 ( 1.1.3 ĸ)
 
  DB2  HP-UX  SCO UnixWare Τĸ Java ʩ
  Java ϯ̪ (UDF)ҳȴȺ̱Ķ Java 
  Ȣ HP-UX  SCO UnixWare ĸ DB2 ջ硨
 
       DB2Stp.java
       DB2Udf.java
       CatUdf.sqlj
       DropUdf.sqlj
       Stp.sqlj
       Udfsrv.java
 
 
| 5.2.2  CLOSE()  SILICON GRAPHICS IRIX ̱
 
|  Silicon Graphics IRIX ĸӬ "close()" ƫȩ
| ϡ롢ȭ桢Ӭ
| ٯ
 
 
  5.2.3  ڷ׻ůΤȢ AND SQLJ Щ SCRIPT
 
  ش Java Ȣġաʩ UDF Ȣšȩ
  Զɱڷ׻ΤȢƫڷ׻ĸϯڷ
  ̽Τĸʩ UDFЩ̯̿Ķ
 
  DB2STP         ܢΤȢƵڷ׻Ȣ DB2SPCLI
 
  DB2UDF         ܢΤȢƵڷ׻Ȣ DB2UDCLI
 
  STP            ܢΤȢƵڷ׻Ȣ STPCLI
 
  UDF            ܢڷ׻ȢƵΤȢ UDFSRV
 
  ǡ Script  "embprep" Щ SQLJ Ȣ
  "makefile" Ƶ Java  SQL Ȣĸ "db2profc"
  թջϯ ID ȴƫſȺɷ
  ڷ׻ڷ̯ DB2 Τĸջ
 
 
  6.0  CLI GUIDE AND REFERENCE (ŧ)
 
 
  6.1  SQLPREPARE()
 
  The description of SQLPrepare() does not indicate that deferred prepare is
  now ON by default.
 
  This means that the processing of the SQL statements passed to the database
  using SQLPrepare() does not take place when the SQLPrepare() function is
  called. Instead, it is processed when SQLExecute() is called.
 
  Minimal error checking is performed by DB2 CLI when SQLPrepare() is called
  (that is, valid function arguments, and so on). The true SQLSTATEs are not
  returned from the prepare until SQLExecute() is called. Your application
  should therefore expect SQLSTATEs from SQLPrepare() after the call to
  SQLExecute(), followed by SQLSTATEs from the SQLExecute() call itself.
 
  For more information see "Deferred Prepare now on by Default" in "Appendix B.
  Migrating Applications", in the section "Changes from Version 2.1.1 to
  5.0.0". This describes how to turn deferred prepare off for Version 2
  programs, as well as a further description of the deferred prepare process.
 
 
  6.2  SQLDISCONNECT()
 
  The description states that cursors are closed by a call to SQLDisconnect().
  This is not the case, however, with stored procedures.
 
  An application should not rely on SQLDisconnect() to close cursors, even if
  it is not a stored procedure. In both cases, the cursor should be closed
  using SQLCloseCursor(), then the statement handle freed with a call to
  SQLFreeHandle() with a HandleType of SQL_HANDLE_STMT.
 
 
  6.3  USING STORED PROCEDURES - PROGRAMMING STORED PROCEDURES TO RETURN
  RESULT SETS
 
  The book indicates that the stored procedures must be run on a remote server.
  This is no longer the case. Stored procedures running on a local server can
  return result sets.
 
  The book indicates that the stored procedure should call SQLFreeStmt() with
  either SQL_DROP or SQL_CLOSE; however, only SQL_CLOSE should be used.
 
 
  6.4  STORED PROCEDURE CATALOG TABLES
 
  All stored procedures should be registered in the new SYSCAT.PROCEDURES and
  SYSCAT.PROCPARMS catalog tables using the CREATE PROCEDURE statement. For
  more information about this SQL statement, see the SQL Reference.
 
  If the stored procedures are not registered using the CREATE PROCEDURE
  statement, CLI or ODBC applications will not know that they exist. For more
  information, see the section "Replacement of the Pseudo Catalog Table for
  Stored Procedures" in the Migration Appendix.
 
 
| 6.5  SQLFETCHSCROLL(SQL_FETCH_RELATIVE, -1)
 
| There is an error in the documentation for SQLFetchScroll() in the version
| 5.0 Call Level Interface Guide and Reference. The table for
| SQL_FETCH_RELATIVE rules should be replaced with the following table:
 
| SQL_FETCH_RELATIVE rules:
 
  +---------------------------------------------------------------------------+
| | Table 3. SQL_FETCH_RELATIVE Rules:                                        |
  +-------------------------------------------+-------------------------------+
| | CONDITION                                 | FIRST ROW OF NEW ROWSET       |
  +-------------------------------------------+-------------------------------+
| | (Before start AND FetchOffset > 0) OR     | -- (a)                        |
| | (After end AND FetchOffset < 0)           |                               |
  +-------------------------------------------+-------------------------------+
| | Before start AND FetchOffset <= 0         | Before start                  |
  +-------------------------------------------+-------------------------------+
| | CurrRowsetStart = 1 AND FetchOffset < 0   | Before start                  |
  +-------------------------------------------+-------------------------------+
| | CurrRowsetStart > 1 AND CurrRowsetStart + | Before start                  |
| | FetchOffset < 1 AND |FetchOffset| >       |                               |
| | RowsetSize                                |                               |
  +-------------------------------------------+-------------------------------+
| | CurrRowsetStart > 1 AND CurrRowsetStart + | 1 (b)                         |
| | FetchOffset < 1 AND |FetchOffset| <=      |                               |
| | RowsetSize                                |                               |
  +-------------------------------------------+-------------------------------+
| | 1 <= CurrRowsetStart + FetchOffset <=     | CurrRowsetStart + FetchOffset |
| | LastResultRow                             |                               |
  +-------------------------------------------+-------------------------------+
| | CurrRowsetStart + FetchOffset >           | After end                     |
| | LastResultRow                             |                               |
  +-------------------------------------------+-------------------------------+
| | After end AND FetchOffset >= 0            | After end                     |
  +-------------------------------------------+-------------------------------+
 
 
  7.0  COMMAND REFERENCE (ŧ)
 
 
  7.1  DB2CMD - OPEN DB2 COMMAND WINDOW
 
  This command is available on Windows NT, Windows 95, and Windows 98.
 
  The following switches are now supported:
 
       /c   Invoke command.exe with the /c option (that is, execute the command,
            and then terminate.  For example, "db2cmd /c dir" causes the "dir" command
            to be invoked in a command window, and then the command window closes.
 
       /w   Wait until the cmd.exe process ends.
            For example, "db2cmd /c /w dir" invokes the "dir" command, and db2cmd.exe
            does not end until the command window closes.
 
       /i   Run the command window, sharing the same console and inheriting file handles.
            For example, "db2cmd /c /w /i db2 get dbm cfg > myoutput"
            invokes cmd.exe to run the db2 command and to wait for its completion.
            A new console is not assigned, and stdout is piped to file "myoutput".
 
       /t   Instead of using "DB2 CLP" as the title of the command window,
            inherit the title from the invoking window. This is useful
            if you want to set up an icon with a different title that
            invokes "db2cmd /t".
 
  NOTE:  All switches must appear before any commands to be executed. For
         example: db2cmd /t db2
 
  If DB21061E ("Command line environment not initialized.") is returned when
  bringing up the CLP-enabled DB2 window, or running CLP commands on Windows 95
  or Windows 98, the operating system may be running out of environment space.
  Check the config.sys file for the SHELL environment setup parameter, and
  increase its value accordingly. For example:
 
       SHELL=C:\COMMAND.COM C:\ /P /E:32768
 
 
  7.2  DB2EVA - EVENT ANALYZER
 
  The "Command Parameters" section is missing the following description:
 
       -evm evmon-name
          Specifies the name of the event monitor whose traces are to be analyzed.
 
 
  7.3  DB2EVMON - EVENT MONITOR PRODUCTIVITY TOOL
 
  The "-db" and "-evm" keywords are case sensitive.
 
 
  7.4  DB2LOOK - DB2 STATISTICS EXTRACTION TOOL
 
  Option "-a" generates statistics for all users on the database. If used with
  option "-e", it generates the DDL for all user tables in the database, but
  statistics are not generated.
 
  NOTES:
 
  1.  Some DDL characteristics are not extracted by db2look.
 
  2.  If neither -u nor -a is specified, the environment variable USER is used.
      On UNIX based systems, this variable does not have to be explicitly set;
     on Windows NT, however, there is no default value for the USER environment
     variable: on this platform, a user variable in the SYSTEM variables must
     be set, or a "set USER=<username>" must be issued for the session.
 
 
  7.5  DB2SET - DB2 PROFILE REGISTRY COMMAND
 
  The "Command Parameters" section has been changed to:
 
       variable=value
          Sets a specified variable to a specified value.
          To delete a variable, do not specify a value for the specified variable.
          Changes to settings take effect after the instance has been restarted.
       -g
          Accesses the global profile variables.
       -i instance
          Specifies the instance profile to use instead of the current, or default.
       node-number
          Specifies a number listed in the db2nodes.cfg file.
       -all
          Displays all occurrences of the local environment variables as defined in:
            o The environment, denoted by [e]
            o The node level registry, denoted by [n]
            o The instance level registry, denoted by [i]
            o The global level registry, denoted by [g].
       -null
          Sets the value of the variable at the specified registry level to null.
          This avoids having to look up the value in the next registry level,
          as defined by the search order.
       -r instance
          Resets the profile registry for the given instance.
       -n DAS node
          Specifies the remote DB2 administration server node name.
       -u user
          Specifies the user ID to use for the administration server attachment.
       -p password
          Specifies the password to use for the administration server attachment.
       -l
          Lists all instance profiles.
       -lr
          Lists all supported registry variables.
       -v
          Specifies verbose mode.
       -h/-?
          Displays help information. When this option is specified,
          all other options are ignored, and only the help information is displayed.
 
 
  7.6  DB2UPD52 - UPDATE CATALOG TO SUPPORT VERSION 5.2
 
  Although this utility is installed at the server, it can be run at a client;
  to do so, copy the executable from the server to the client.
 
  In an MPP environment, the utility must be run from the catalog node of the
  specified database.
 
 
  7.7  BACKUP DATABASE
 
  To use tape devices, DB2 users on SCO UnixWare 7 need to specify
  BUFFER to be 16. The default value of BUFFER is 1024 pages. If BUFFER is set
  to zero, the database manager configuration parameter BACKBUFSZ must be set
  to 16.
 
 
  7.8  EXPORT
 
  Table 6 ("Valid Delimiters"), and all references to it, should be removed.
  The information contained in the note at the bottom of the table, and the restrictions
  that follow the note, are correct, however.
 
  NOTE:  In a DBCS environment, the pipe (|) character delimiter is not supported.
 
  Replace the section "DB2 Data Links Manager Considerations" in the Usage
  Notes with the following:
 
  To ensure that a consistent copy of the table and the corresponding files
  referenced by the DATALINK columns are copied for export, do the following:
 
 
  1.  To ensure that no update transactions are in progress when the export
      utility is run, issue the command QUIESCE TABLESPACES FOR TABLE tablename
      SHARE.
 
  2.  Issue the EXPORT command.
 
  3.  Run the dlfm_export utility with root authority at each file server.
      Input to the dlfm_export utility is the control file, server_name, which
      is generated by the export utility. The dlfm_export utility must be run
      with root authority in order to successfully archive files to which the
      DLFM administrator may not have access.
 
  4.  To make the table available for updates, issue the command QUIESCE
      TABLESPACES FOR TABLE tablename RESET.
 
  EXPORT is executed as an SQL application. The rows and columns satisfying the
  SELECT statement conditions are extracted from the database. For the DATALINK
  columns, the SELECT statement should not specify any scalar function.  The
  export utility uses APIs to extract parts of the DATALINK value, such as link
  type, file server name, file path name, and comments.
 
  Successful execution of EXPORT results in generation of the following files:
 
  o   An export data file as specified in the EXPORT command. A DATALINK column
      value in this file is in the format described on page 290. When the
      DATALINK column value is the SQL NULL value, handling is the same as that
      for other data types.
 
  o   Control files server_name, which are generated for each file server (on
      the Windows operating system and on OS/2, a single control file,
      ctrlfile.lst, is used by all the file servers). A control file contains
      the URLs for all the files that are to be exported from that file server.
 
  Use the dlfm_export utility to export files from a file server as follows:
 
        Usage: dlfm_export <control file name> [<tar file name>]
        Description:
          <control file name>  This is the file name that was generated by
                               running the export utility on the DB2 client.
          <tar file name>      This is the name of the archive file that will
                               be generated (the default is export.tar
                               in the current working directory).
 
  A corresponding utility called dlfm_import is provided to retrieve and to
  restore files from the archive that dlfm_export generates. This utility must
  be used whether the archived files are being restored on the same or on a
  different file server. Use the dlfm_import utility to retrieve files from the
  archive as follows:
 
        Usage: dlfm_import [<tar file name>]
        Description:
          <tar file name>      This is the name of the archive file that will
                               be used to recover the files (the default is export.tar
                               in the current working directory).
 
  NOTES:
 
  1.  Both dlfm_export and dlfm_import utilities must be run with root
      authority. In the case of dlfm_export, there may be files to be archived
      to which the DLFM administrator does not have access. In the case of
      dlfm_import, root authority is required because the user may want to
      restore the archived files on a different file server, one that may not
      have the same directory structure and user IDs as the file server on
      which the dlfm_export utility was run.
 
  2.  The File Manager does not have to be running in order for the dlfm_export
      and the dlfm_import utilities to run.
 
  3.  When running the dlfm_import utility on a file server other than the one
      on which dlfm_export was run, the files will be restored on the correct
      paths. The files will be owned by root, in case some of the user IDs do
      not exist on the new machine. Before inserting the corresponding DATALINK
      values into the database, it is the administrator's responsibility to
      ensure that all files have the correct permissions and belong to the
      correct user IDs.
 
  EXPORTING BETWEEN INSTANCES
 
  The table below shows how to export DB2 data and the files that are
  referenced by the instance called SystemA to the instance called SystemB.
  SystemA uses file servers DLFM1 and DLFM2. SystemB uses file servers DLFMX
  and DLFMY. The files on DLFM1 will be exported to DLFMX, and the files on
  DLFM2 will be exported to DLFMY.
 
    +-----------------------------------------------------------------------------+
    | InstanceA with File Servers DLFM1 and DLFM2  | Step                          |
    +------------------------------------------------------------------------------+
    | DB2 Data | File1 for DLFM1 | File2 for DLFM2 |                               |
    | on File  |                 |                 |                               |
    +----------+-----------------+-----------------+-------------------------------+
    |          | Run the         | Run the         | 1.Run dlfm_export (as root)  |
    |          | dlfm_export     | dlfm_export     | on both file servers.        |
    |          | command         | command         | This will produce an archive  |
    |          |                 |                 | on both file servers.        |
    +----------+-----------------+-----------------+-------------------------------+
 
    +-----------------------------------------------------------------------------+
    | InstanceB with File Servers DLFMX and DLFMY  | Step                          |
    +------------------------------------------------------------------------------+
    |          | On DLFMX,       | On DLFMY,       | 2. Run dlfm_import (as root) |
    |          | restore from an | restore from an | on both file servers.        |
    |          | archive         | archive         |                               |
    +----------+-----------------+-----------------+-------------------------------+
    |          |                 |                 | 3. Run the IMPORT command on |
    |          |                 |                 | InstanceB, with the parameter |
    |          |                 |                 | DL_URL_REPLACE_PREFIX to      |
    |          |                 |                 | specify the appropriate file  |
    |          |                 |                 | server for each exported file.|
    +----------+-----------------+-----------------+-------------------------------+
 
    +-----------------------------------------------------------------------------+
    | After the import utility is run on InstanceB, InstanceA data and all         |
    | files referenced by DATALINK columns are imported.                          |
    +------------------------------------------------------------------------------+
 
 
  7.9  GET DATABASE MANAGER CONFIGURATION
 
 
  7.9.1  NEW PARAMETER CATALOG_NOAUTH
 
  A new parameter, catalog_noauth, specifies whether users are able to catalog
  and uncatalog databases and nodes, or DCS and ODBC directories, without
  SYSADM authority. The default value (0) for this parameter indicates that
  SYSADM authority is required. When this parameter is set to 1 (yes), SYSADM
  authority is not required.
 
 
| 7.9.2  TRUST_ALLCLNTS UPDATED
 
| "TRUST_ALLCLNTS" as well as "TRUST_CLNTAUTH" are used to determine where
| users are validated for the database environment. By accepting the default
| for "TRUST_ALLCLNTS", all clients are treated as trusted clients. This means
| a level of security is available at the client, and that users can be
| validated at the client. Other options may be used to protect the server
| against certain clients based on their platform or database protocol.
 
 
  7.10  IMPORT
 
  In an MPP environment, if a database was bound using the INSERT BUF option,
  buffered insert will be disabled during any import operation to that database
  if the INSERT_UPDATE is specified.
 
  NOCHECKLENGTHS is supported for both the ASC and the IXF file formats. This
  option does not have to be used with the FORCEIN modifier. If NOCHECKLENGTHS
  is specified, an attempt is made to import each row, even if the source data
  has a column definition that exceeds the size of the target table column.
  Such rows may be successfully imported if code page conversion causes the
  source data to shrink; for example, 4-byte EUC data in the source could
  shrink to 2-byte DBCS data in the target, and require half the space. This
  option is particularly useful if it is known that the source data will fit in
  all cases despite mismatched column definitions.
 
  The current default priority for delimiters is: record delimiter,
  character delimiter, column delimiter. To protect existing customer applications that
  depend on the older priority, a new file type modifier, DELPRIORITYCHAR, has
  been created for the import utility. This modifier reverts the delimiter
  priorities to: character delimiter, record delimiter, column delimiter. The
  syntax is: db2 import ... modified by delprioritychar ...
 
 
  7.11  LIST DATABASE DIRECTORY
 
  There can be a maximum of eight opened database directory scans per process.
  To overcome this restriction for a batch file that issues more than eight
  LIST DATABASE DIRECTORY commands within a single DB2 session, convert the
  batch file into a shell script. The "db2" prefix generates a new DB2 session
  for each command.
 
 
  7.12  LOAD
 
  This command affects only the partition to which a direct connection exists;
  the load utility operates on a single database partition only.
 
  The COPY option is not supported for tables with DATALINK columns.
 
  The REPLACE option is not supported for tables with DATALINK columns.
 
  The NONRECOVERABLE clause should not be used when DATALINK columns with the
  FILE LINK CONTROL attribute are present in - or being added to - the table.
 
  There is a limit on the combination of DATALINK columns and the number of
  DLFMs used. There can be at most 256 unique combinations of DLFMs and
  DATALINK columns in a table. For example, the table can have a maximum
  of 128 DATALINK columns if the number of DLFMs configured is two.
 
  Load copy to tapes is not supported for DB2 servers running on SCO UnixWare 7.
 
  The Usage Notes section incorrectly states that "summary tables" that are
  dependent on the tables being loaded are placed in check pending state. It
  should state that "summary tables defined with REFRESH IMMEDIATE", that are
  dependent on the tables being loaded, are placed in check pending state.
 
  NOCHECKLENGTHS is supported for both the ASC and the IXF file formats. This
  option does not have to be used with the FORCEIN modifier. If NOCHECKLENGTHS
  is specified, an attempt is made to load each row, even if the source data
  has a column definition that exceeds the size of the target table column.
  Such rows may be successfully loaded if code page conversion causes the
  source data to shrink; for example, 4-byte EUC data in the source could
  shrink to 2-byte DBCS data in the target, and require half the space. This
  option is particularly useful if it is known that the source data will fit in
  all cases despite mismatched column definitions.
 
  The current default priority for delimiters is: record delimiter,
  character delimiter, column delimiter. To protect existing customer applications
  that depend on the older priority, a new file type modifier, DELPRIORITYCHAR,
  has been created for the load utility. This modifier reverts the delimiter
  priorities to: character delimiter, record delimiter, column delimiter. The
  syntax is: db2 load ... modified by delprioritychar ... The new option is not
  supported in an MPP environment, nor is it supported by the AutoLoader. Load
  SMP exploitation is disabled when this option is specified (CPU_PARALLELISM
  is forced to a value of 1). The new option is mutually exclusive
  with the "DUMPFILE" feature, and is supported only for the DEL file format.
 
 
  7.13  RECONCILE
 
  There is an error in the description of "DLREPORT filename". There is no
  default value for this parameter.
 
  The file name must be a fully qualified file name. For example:
 
       /home/user/report.out
 
 
  7.14  RESTORE DATABASE
 
  If the WITHOUT DATALINK option is not specified, and the DB2 File Manager
  containing the DATALINK data is unavailable, the restore operation will fail.
 
  If this option is specified, and the DB2 File Manager containing the DATALINK
  data is unavailable, all table spaces which contain tables with DATALINK
  values on the unavailable server are placed in RESTORE PENDING state.
 
  To use tape devices, DB2 users on SCO UnixWare 7 need to specify
  BUFFER to be 16. The default value of BUFFER is 1024 pages. If BUFFER is set
  to zero, the database manager configuration parameter BACKBUFSZ must be set
  to 16.
 
 
| 7.15  START DATABASE MANAGER
 
| When running the "db2start" command (or using the "NET START" command) to
| start the database manager on the Windows NT environment, the command will
| not return any warnings if any communication subsystem failed to start. The
| user should always examine the NT Event Log or the "DB2DIAG.LOG" for any
| errors that may have occurred during the running of "db2start". This occurs
| because the database manager on a Windows NT environment is implemented as an
| NT service, and hence does not return an error if the service is started
| successfully.
 
 
  8.0  EMBEDDED SQL PROGRAMMING GUIDE (ŧ)
 
 
  8.1  WRITING USER-DEFINED FUNCTIONS (INTERFACE BETWEEN DB2 AND A UDF/THE
  ARGUMENTS PASSED FROM DB2 TO A UDF)
 
  There are two changes to the numbered list under the "dbinfo" argument:
 
  o   Insert a new item 17 in the list. The new item 17 reads as follows:
 
           17. Unique application identifier (appl_id)
 
              This field is a pointer to a C null-terminated string which
              uniquely identifies the application's connection to DB2.
              It is regenerated at each database connect.
 
              The string has a maximum length of 32 characters, and its
              exact format depends on the type of connection established
              between the client and DB2.  Generally it takes the form
 
                <x>.<y>.<ts>
 
              where the <x> and <y> vary by connection type, but the <ts>
              is a 12 character time stamp of the form YYMMDDHHMMSS, which
              is potentially adjusted by DB2 to ensure uniqueness.
 
              Example:  *LOCAL.db2inst.980707130144
 
  o   Change the existing item 17 to item 18, and within this item, change "24
      characters" to "20 characters":
 
           17. Reserved field (resd2)
 
              This field is for future use. It is defined as 20 characters long.
 
 
  8.2  PROGRAMMING IN C AND C++ (PROGRAMMING CONSIDERATIONS/HOST VARIABLES)
 
 
  8.2.1  DECLARING HOST VARIABLES
 
  BIGINTs can now be declared in one of three ways:
 
  o   long long
 
  o   __int64 // (A double underscore precedes "int64".)
 
      This declaration is specific to the Microsoft compiler.
 
  o   sqlint64
 
      This is useful when writing platform-independent applications. The DB2
      UDB sqlsystm.h header file will type define sqlint64 as "__int64" on the
      Windows NT platform when using the Microsoft compiler; otherwise,
      sqlint64 will be type defined as "long long".
 
 
  8.2.2  HANDLING GRAPHIC HOST VARIABLES (WINDOWS OPERATING SYSTEM)
 
  When specifying the WCHARTYPE CONVERT option on a Windows platform, you should note
  that wchar_t on Windows platforms is Unicode. Therefore, if your C/C++ compiler's
  wchar_t is not Unicode, the wcstombs() function call may fail with SQLCODE -1421
  (SQLSTATE=22504). If this happens, you can specify the WCHARTYPE NOCONVERT option,
  and explicitly call the wcstombs() and mbstowcs() functions from within your program.
 
 
| 8.3  PROGRAMMING IN COBOL (HOST STRUCTURE SUPPORT)
 
| The COBOL precompiler supports declarations of group data items in the host
| variable declare section. Among other things, this provides a shorthand for
| referring to a set of elementary data items in an SQL statement. For example,
| the following group data item can be used to access some of the columns in
| the STAFF table of the SAMPLE database:
 
|          01 staff-record.
|              05 staff-id       pic s9(4) comp-5.
|              05 staff-name.
|                  49 l          pic s9(4) comp-5.
|                  49 d          pic x(9).
|              05 staff-info.
|                  10 staff-dept pic s9(4) comp-5.
|                  10 staff-job  pic x(5).
 
| Group data items in the declare section can have any of the valid host
| variable types described above as subordinate data items. This includes all
| numeric and character types, as well as all large object types. THE
| RESTRICTION OF ALLOWING ONLY ONE LEVEL OF GROUP DATA ITEMS HAS BEEN REMOVED,
| AND THEY CAN NOW BE NESTED UP TO 10 LEVELS. Note that you must declare
| VARCHAR character types with the subordinate items at level 49, as in the
| above example. If they are not at level 49, the VARCHAR is treated as a group
| data item with two subordinates, and is subject to the rules of declaring and
| using group data items. In the example above, "staff-info" is a group data
| item, whereas "staff-name" is a VARCHAR. The same principle applies to LONG
| VARCHAR, VARGRAPHIC and LONG VARGRAPHIC. You may declare group data items at
| any level between 02 and 49.
 
| You can use group data items and their subordinates in four ways:
 
| METHOD 1.
 
| The entire group may be referenced as a single host variable in an SQL
| statement:
 
|             EXEC SQL SELECT id, name, dept, job
|               INTO :staff-record
|               FROM staff WHERE id = 10 END-EXEC.
 
| The precompiler converts the reference to staff-record into a list, separated
| by commas, of all the subordinate items declared within staff-record. Each
| elementary item is qualified with the group names of all levels to prevent
| naming conflicts with other items. This is equivalent to the following
| method.
 
| METHOD 2.
 
| The second way of using group data items:
 
|             EXEC SQL SELECT id, name, dept, job
|               INTO
|               :staff-record.staff-id,
|               :staff-record.staff-name,
|               :staff-record.staff-info.staff-dept,
|               :staff-record.staff-info.staff-job
|               FROM staff WHERE id = 10 END-EXEC.
 
| NOTE:  The reference to "staff-id" is qualified with its group name using the
|        prefix "staff-record.", and not "staff-id" of "staff-record" as in
|        pure COBOL.
 
| Assuming there are no other host variables with the same names as the
| subordinates of staff-record, the above statement can also be coded as in
| method 3, eliminating the explicit group qualification.
 
| METHOD 3.
 
| Here, subordinate items are referenced in a typical COBOL fashion, without
| being qualified to their particular group item:
 
|             EXEC SQL SELECT id, name, dept, job
|               INTO
|               :staff-id,
|               :staff-name,
|               :staff-dept,
|               :staff-job
|               FROM staff WHERE id = 10 END-EXEC.
 
| As in pure COBOL, this method is acceptable to the precompiler as long as a
| given subordinate item can be uniquely identified. If, for example,
| "staff-job" occurs in more than one group, the precompiler issues an error
| indicating an ambiguous reference:
 
|             SQL0088N Host variable "staff-job" is ambiguous.
 
| METHOD 4.
 
| To resolve the ambiguous reference, partial qualification of the subordinate
| item can also be used.
 
|             EXEC SQL SELECT id, name, dept, job
|               INTO
|               :staff-id,
|               :staff-name,
|               :staff-info.staff-dept,
|               :staff-info.staff-job
|               FROM staff WHERE id = 10 END-EXEC.
 
| Because a reference to a group item alone, as in method 1, is equivalent to a
| comma-separated list of its subordinates, there are instances where this type
| of reference leads to an error. For example:
 
|             EXEC SQL CONNECT TO :staff-record END-EXEC.
 
| Here, the CONNECT statement expects a single character-based host variable.
| By giving the "staff-record" group data item instead, the host variable
| results in the following precompile-time error:
 
|             SQL0087N Host variable "staff-record" is a structure used where
|                      structure references are not permitted.
 
| Other uses of group items which cause an SQL0087N to occur include PREPARE,
| EXECUTE IMMEDIATE, CALL, indicator variables, and SQLDA references.  Groups
| with only one subordinate are permitted in such situations, as are references
| to individual subordinates, as in method 2, 3 and 4 above.
 
| INDICATOR TABLES
 
| The COBOL precompiler supports the declaration of tables of indicator
| variables, which are convenient to use with group data items. They are
| declared as follows:
 
|        01 <indicator-table-name>.
|            05 <indicator-name> pic s9(4) comp-5
|                                occurs <table-size> times.
 
| For example:
 
|        01 staff-indicator-table.
|            05 staff-indicator pic s9(4) comp-5
|                               occurs 7 times.
 
| This indicator table can be used effectively with the first format of group
| item reference above:
 
|         EXEC SQL SELECT id, name, dept, job
|           INTO :staff-record :staff-indicator
|           FROM staff WHERE id = 10 END-EXEC.
 
| Here, the precompiler detects that "staff-indicator" was declared as an
| indicator table, and expands it into individual indicator references when it
| processes the SQL statement. "staff-indicator"(1) is associated with
| "staff-id" of "staff-record", "staff-indicator"(2) is associated with
| "staff-name" of "staff-record", and so on.
 
| NOTE:  If there are "k" more indicator entries in the indicator table than
|        there are subordinates in the data item (for example, if
|        "staff-indicator" has 10 entries, making "k"=6), the "k" extra entries
|        at the end of the indicator table are ignored.  Likewise, if there are
|        "k" fewer indicator entries than subordinates, the last "k"
|        subordinates in the group item do not have indicators associated with
|        them. NOTE THAT YOU CAN REFER TO INDIVIDUAL ELEMENTS IN AN INDICATOR
|        TABLE IN AN SQL STATEMENT.
 
 
  8.4  PROGRAMMING IN JAVA (CREATING JAVA UDFS AND STORED PROCEDURES/CLASSES
  FOR JAVA STORED PROCEDURES AND UDFS)
 
  In the section "COM.ibm.db2.app.UDF", there is a list of functions that
  return specific DBINFO fields. To this list has been added:
 
       public String getDBapplid() throws Exception
 
 
  8.5  A JDK PROBLEM THAT AFFECTS JDBC PROGRAMS RUNNING UNDER A DBCS CODE PAGE
  ON AIX
 
  When running JDBC programs with JDK 1.1.4 bundled with AIX 4.3.1 under a DBCS
  code page, the DBCS data can not be displayed properly. This is a known
  problem in this version of JDK, and a fix for it (IX78932) is available.
 
 
  8.6  JDBC SAMPLE APPLICATIONS RESTRICTION ON SCO UNIXWARE 7
 
  Because a native method cannot make use of the thread library (libthread.so
  from UnixWare 7; see the UnixWare release notes), DB2 JDBC sample
  applications (DB2Appl, DB2Udf, and DB2Stp) do not work from a Version
  5.2 client on UnixWare 7 connecting to a Version 5 UDB server, or a
  Version 5 client connecting to a FixPak 3 UDB server.
 
 
| 8.7  JDBC 2.0
 
| JDBC 2.0 is the latest version of JDBC from Sun. This version of JDBC has two
| defined parts: the CORE API, and the STANDARD EXTENSION API. The core API has
| been released. For information on its specification, refer to:
 
|      http://www.software.ibm.com/data/db2/java
 
| The standard extension API is currently under review. For information on the
| review version of its specification, you can also visit the web page above.
 
| FEATURES
 
| The JDBC 2.0 driver we ship in this FixPak supports the JDBC 2.0 core API.
| However, due to the unavailability of certain features of the DB2 Engine, not
| all features defined in the specification are supported. Here is a list of
| main features that are supported:
 
| o   Read-only Scrollable Result Set
 
| o   Batch Update (Batch statements and batch prepared statements)
 
| o   LOB support
 
| Here is a list of features that are NOT supported:
 
| o   Updatable Scrollable ResultSet
 
| o   New SQL types (Array, Ref, Distinct, Java Object)
 
| o   Customized SQL type mapping
 
| PREREQUISITE
 
| JDK 1.2
 
| PLATFORM
 
| The JDBC 2.0 driver is currently available only on Windows 95, Windows 98,
| and Windows NT operating systems.
 
| COMPATIBILITY
 
| This version of the specification is backward compatible with the previous
| version (1.22). However, the DB2 JDBC 1.22 driver supports LOB types as an
| extension of the JDBC 1.22 specification, and this extension is not part of
| the new specification's backward compatibility. This means that existing JDBC
| applications that rely on the LOB support of the JDBC 1.22 driver may not
| work with the new driver. To fix the problem, we recommend that you modify
| the application to be compliant with the new specification. However, this
| solution may not be practical for every situation. As a workaround we provide
| a keyword, "JDBCVERSION", that you can set to "122" to tell the JDBC driver
| you want the 1.22 version for DB2 LOB support. The default is "200" for 2.0
| version behavior. You can set this keyword in "db2cli.ini", or pass it in as
| a connection attribute in the "getConnection" property argument.
 
| INSTALLATION
 
| In this FixPak, the JDBC 1.22 driver is still the default driver. The JDBC
| 2.0 driver is included in the "sqllib\java12" directory. To use the JDBC 2.0
| driver, run the "usejdbc2" batch file. This will create an "sqllib\java11"
| directory for the 1.22 driver files. It will also back up the JDBC 1.22
| driver files into this directory, and will copy the JDBC 2.0 driver files
| from the "sqllib\java12" directory into the appropriate directories. To
| switch back to the JDBC 1.22 driver, execute the "usejdbc1" batch file.
 
 
| 8.8  PERL ACCESS TO DB2 UDB DATABASES
 
| Release 0.70 of the DB2 UDB driver (DBD::DB2) for the Perl Database Interface
| (Perl DBI) is now available.
 
| The Perl DBI is an Application Programming Interface (API) that provides
| database access for the Perl language. It defines a set of functions,
| variables, and conventions that provide a consistent database interface
| independent of the actual database being used.
 
| The DBD::DB2 driver works in conjunction with the DBI to access DB2 UDB.
 
| For additional information, see the Web page:
 
|      http://www.software.ibm.com/data/db2/perl/
 
 
  8.9  DATALINK DATA TYPE
 
  Embedded SQL using the DATALINK data type is not supported.
 
 
  8.10  MULTITHREADED UNIX APPLICATIONS WORKING WITH CODE PAGE AND COUNTRY CODE
 
  On AIX, Solaris, HP-UX, SCO UnixWare 7, and Silicon Graphics IRIX , changes
  have been made to the functions that are used for run time querying of the
  code page and country code to be used for a database connection. They are now
  thread safe but can create some lock contention (and resulting performance
  degradation) in a multithreaded application which uses a large number of
  concurrent database connections.
 
  A new environment variable has been created (DB2_FORCE_NLS_CACHE) to
  eliminate the chance of lock contention in multithreaded applications. When
  DB2_FORCE_NLS_CACHE is set to TRUE the code page and country code information
  is saved the first time a thread accesses it. From that point on the cached
  information will be used for any other thread that requests this information.
  By saving this information, lock contention is eliminated and in certain
  situations a performance benefit will be realized.
 
  DB2_FORCE_NLS_CACHE should not be set to true if the application changes
  locale settings between connections. If this is done then the original locale
  information will be returned even after the locale settings have been
  changed. In general, multithreaded applications will not change locale
  settings. This ensures that the application remains thread safe.
 
 
  8.11  ERROR PROCESSING FOR UDF CALL TYPES
 
  8.11.1  SCRATCHPAD AND FINAL CALL TYPE ARGUMENTS FOR EXTERNAL TABLE FUNCTIONS
 
  Changes have been introduced (a) to give the user explicit control over
  scratchpad duration, and (b) to avoid inconsistencies in the duration of
  scratchpads, for table functions written in:
 
  o   C/C++ - until now, scratchpads always survived across CLOSE/OPEN calls to
      the UDF
 
  o   OLE/Java - until now, a fresh scratchpad was always allocated on OPEN
 
  This difference has led to the possibility of getting inconsistent results,
  for example when performing a join.
 
  The following changes have been introduced for User Defined Functions (UDFs)
  which are table functions:
 
  1.  FINAL CALL is now optional for the CREATE FUNCTION statement for table
      functions, as it is for scalar functions. NO FINAL CALL (the default) may
      thus be specified for table functions.
 
  2.  When writing a table function which has a scratchpad, you can elect to
      have the scratchpad be initialized for each OPEN call to the function, by
      specifying NO FINAL CALL, in which case the table function is expected to
      release resources on each CLOSE call.
 
      Or, you can elect to have the scratchpad content be preserved across OPEN
      calls, by specifying FINAL CALL. If you do this, you need not release
      resources during the CLOSE call processing, because a FINAL call will be
      made to the table function at end-of-statement. This FINAL CALL is new
      with this change, as is the balancing FIRST call which takes place before
      the first OPEN call. These two new call types occur only on table
      functions which specify FINAL CALL.
 
  3.  For Java table functions, a change to the execution model needs to be
      understood. The following table explains what happens at each significant
      point in the table function processing of a given statement (the bottom
      part of each box hints what the code might be written to do for a typical
      table function which pulls some information in from the Web). Covered are
      both the NO FINAL CALL and the FINAL CALL cases, assuming SCRATCHPAD in
      both cases, which account for the new rules described above.
 
                      |                              |
      Point     | |    NO FINAL CALL             |    FINAL CALL
      in        | |    LANGUAGE JAVA             |    LANGUAGE JAVA
      scan      | |    SCRATCHPAD                |    SCRATCHPAD
      time      V |                              |
          ------------+------------------------------+------------------------------
      Before the  | No calls.                    | Class constructor is called
      first OPEN  |                              |  (means new scratchpad).
          for the     |                              | UDF Method is called with
      table func. |                              |  "FIRST" call.
                      |                              |- - - - - - - - - - - - - - -
                  |                              | Constructor initializes
                  |                              |  class and scratchpad
                  |                              |  variables.
                      |                              | Method connects to the Web
                  |                              |  Server.
                      |                              |
          ------------+------------------------------+------------------------------
      At each     | Class constructor is called  | UDF Method is called with
      OPEN of the |  (means new scratchpad).     |  "OPEN" call.
          table func. | UDF Method is called with    |
                  |  "OPEN" call.                |
                  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
                  | Constructor initializes      | Method opens the scan for
                  |  class and scratchpad        |  whatever Web data it wants.
                      |  variables.                  |  (Might be able to avoid
                  | Method connects to the Web   |  reopen after a CLOSE repo-
                  |  Server, and opens the scan  |  sition, depending on what
                  |  for Web data.               |  is saved in scratchpad!)
                  |                              |
          ------------+------------------------------+------------------------------
      At each     | UDF method is called with    | UDF Method is called with
      FETCH for   |  "FETCH" call.               |  "FETCH" call.
          new row of  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
      table func  | Method fetches and returns   | Method fetches and returns
      data.       |  next row of data, or EOT.   |  next row of data, or EOT.
                      |                              |
          ------------+------------------------------+------------------------------
      At each     | UDF method is called with    | UDF Method is called with
      CLOSE of the|  "CLOSE" call.               |  "CLOSE" call.
          table func. | "close" method is called if  |
                  |  it exists for class.        |
                  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
                  | Method closes its Web scan   | Method might reposition to
                  |  and disconnects from the    |  the top of the scan, or
                  |  Web Server.                 |  close the scan.  It can
                  | "close" does not need to do  |  save any state in the
                  |  anything.                   |  scratchpad, which will
                  |                              |  persist.
                      |                              |
          ------------+------------------------------+------------------------------
      After the   | No calls.                    | UDF Method is called with
      last CLOSE  |                              |  "FINAL" call.
          of the table|                              | "close" method is called if
      func.       |                              |  it exists for class.
                      |                              |- - - - - - - - - - - - - - -
                  |                              | Method disconnects from the
                  |                              |  Web Server.
                      |                              | "close" does not need to do
                  |                              |  anything.
                      |                              |
          ------------+------------------------------+------------------------------
 
      Some notes on this table:
 
      a.  By "UDF Method" we mean the Java class method which implements the
          UDF; the method identified in the EXTERNAL NAME clause of the CREATE
          FUNCTION statement.
 
      b.  For table functions with NO SCRATCHPAD specified, the calls to the UDF
      method are as indicated in this table, but because the user is not asking
      for any continuity via a scratchpad, DB2 will cause a new object to be
      instantiated before each call, by calling the class constructor. It is
      not clear that table functions with NO SCRATCHPAD (and thus no
      continuity) can do very useful things, but they are supported.
 
      c.  These execution models imply the existence of a new class method
          which provides the call type. This method is provided, as are
          numerous other methods, with DB2 in the COM.ibm.db2.app.UDF class,
          and will be defined as:
 
               public int getCallType () throws Exception
 
      d.  These models are TOTALLY COMPATIBLE with what happens with the other
          UDF languages: C and C++ and OLE.
 
  Defines for the new call types are added to sqludf.h, the UDF include file.
 
  UDB customers who have existing table functions are advised to examine the
  above options and make appropriate changes:
 
  1.  If you depend on the new scratchpad rules implied by FINAL CALL then you
      may have to change your UDF to add the new call types, because with FINAL
      CALL specified, the new calls will be made to the UDF. Also, any
      acquired resources should be freed when the FINAL call is handled by the
      table function.
 
  2.  If you wish to use the rules implied by NO FINAL CALL, you are advised to
      change and rerun your CREATE FUNCTION statement with NO FINAL CALL
      instead of FINAL CALL, so that your UDF will not get the new calls.
      (Depending on how the UDF is written to examine the call-type argument,
      it may or may not give bad results if it receives unexpected calls.)
 
  3.  If you have a Java table function, then you will have to change your
      function to conform to the new execution models in the above table. For
      either execution model (depending on whether you choose FINAL CALL or NO
      FINAL CALL), you will note that additional calls are made to the UDF
      method. This gives additional power to the UDF writer, but means that he
      or she must distinguish the calls. Use of the new getCallType method in
      the parent class included by DB2 provides this capability.
 
  Detailed information changes will appear in the "SQL Reference" and the
  "Embedded SQL Programming Guide" (ESPG) when they are next updated.
  Meanwhile, a "Technote" describing these changes in more detail will be made
  available in the DB2 Technical Library on the World Wide Web. Use the search
  argument "FINAL CALL".
 
  The following sections describe the detailed changes to the "Embedded SQL
  Programming Guide".
 
  In Chapter 6 of the "Embedded
  SQL Programming Guide", under the heading "User-Defined Functions (UDF)", in
  the Example "Table Function Returning Document IDs", "FINAL CALL" should now
  read "NO FINAL CALL", and in the following paragraph note that FINAL CALL
  need not be specified for each table function.
 
  In Chapter 7 of the "Embedded SQL Programming Guide", under the heading
  "Interface between DB2 and a UDF", in the section "The Arguments Passed from
  DB2 to a UDF", in the description for the "scratchpad" argument, the text
  beginning "The actual..." should be replaced by:
 
  For table functions, the scratchpad is initialized as above only for the very
  first call to the UDF if FINAL CALL is specified on the CREATE FUNCTION.
  This "very first call" has type FIRST. After this call, the scratchpad
  content is totally under control of the table function.
 
  If NO FINAL CALL was specified or defaulted, then the scratchpad is
  initialized as above for each OPEN call, and the scratchpad content is
  completely under control of the table function between OPEN calls. (This can
  be very important for a table function used in a join or subquery. If it is
  necessary to maintain the content of the scratchpad across OPEN calls, then
  FINAL CALL must be specified in your CREATE FUNCTION statement. With FINAL
  CALL specified, in addition to the normal OPEN, FETCH and CLOSE calls, the
  table function will also receive FIRST and FINAL calls, for the purpose of
  scratchpad maintenance and resource release.)
 
  Also, in the description of the "call-type" argument, the second sentence
  should be replaced by the following sentence:
 
  For external scalar functions this argument is only present if FINAL CALL is
  specified in the CREATE FUNCTION statement, but for external table functions
  it is ALWAYS present.
 
  After that first paragraph, add the following new paragraph:
 
  Note that even though all the current possible values are listed below, your
  UDF should contain a switch or case statement which explicitly tests for all
  the expected values, rather than containing "if A do AA, else if B do BB,
  else it must be C so do CC" type logic. This is because it is possible that
  additional call types may be added in the future, and if you don't explicitly
  test for condition C you will have trouble when new possibilities are added.
 
  In the same section, where the values for the "call-type" argument are
  explained for Table Functions, the text should be amended as follows:
 
  o   -2 This is the FIRST call, which only occurs if the FINAL CALL keyword
      was specified for the UDF. The scratchpad is set to binary zeros before
      this call.  Argument values are passed to the table function, and it may
      choose to acquire memory or perform other one-time only resource
      initialization. Note that this is not an OPEN call, that call follows
      this one. On a FIRST call the table function should not return any data
      to DB2, but it can set SQL-state and diagnostic-message argument return
      values.
 
  o   -1 This is the OPEN call. The scratchpad will be initialized if NO FINAL
      CALL is specified, but not necessarily otherwise. (The rest of the text
      is unchanged.)
 
  o   0 (unchanged)
 
  o   1 (The first paragraph is unchanged, but the following text should be
      added) During CLOSE, the UDF should release resources acquired, if NO
      FINAL CALL was defined for the UDF. If FINAL CALL was defined, a special
      call (FINAL) is provided for resource release. On a CLOSE call a table
      function should not return a row to DB2, and in fact any SQL-result and
      SQL-result-ind values returned are ignored by DB2 upon return. However,
      the table function may set SQL-state and diagnostic-message argument
      values, which are handled as usual by DB2.
 
  o   2 This is the FINAL call, which balances the FIRST call, and it only
      occurs if FINAL CALL was defined for the external table function. All
      system resources should be freed. This call only occurs once per
      statement. As usual, SQL-state and diagnostic-message argument values may
      be returned, and they are handled as usual by DB2.
 
  Under the heading "Summary of Argument Use", for the argument "scratchpad and
  call-type", add the text ", and optionally FIRST and FINAL" in both cases.
 
  In Chapter 7 of the "Embedded SQL Programming Guide", under the heading
  "Table Function Considerations", under the first bullet, there are now five
  call types instead of three: add FIRST and FINAL to the list.
 
  Also, add a new bullet in this section as well, which should read as follows:
 
  The error processing model for table function calls is as follows:
 
  1.  If FIRST call fails, no further calls are made.
 
  2.  If FIRST call succeeds, FINAL call is made.
 
  3.  If OPEN call fails, no FETCH or CLOSE call is made.
 
  4.  If OPEN call succeeds, then FETCH and CLOSE calls are made.
 
  In the "Weather Table Function" example in Chapter 7, the UDF definition
  should specify NO FINAL CALL, and note that FINAL CALL is no longer mandatory
  for a table function (that is, ignore the statement that says it is
  mandatory).
 
  In Chapter 15 of the "Embedded SQL Programming Guide", entitled "Programming
  in Java", under the heading "Creating Java UDFs and Stored Procedures", the
  second sentence of the second paragraph should be changed to read as follows:
 
  Once you create and register these UDFs and stored procedures, and place the
  Java classes in the correct file location, described in "Where to Put Java
  Classes" on page 514, you can then (for UDFs) reference them in your SQL, or
  (for stored procedures) call them from your application written in any
  supported language.
 
  And the sentence following that one should be removed.
 
  In Chapter 15 of the "Embedded SQL Programming Guide", under the heading
  "Creating and Using Java User-Defined Functions", the fourth sentence of the
  first paragraph should be changed to read:
 
  You can then refer to it in the SQL of your application.
 
  In the subheading "Changing How a Java UDF Runs" under that same heading, the
  first three paragraphs should be replaced with the following five:
 
  Typically, DB2 calls a UDF many times, once for each row of an input or
  result set in a query. If SCRATCHPAD is specified in the CREATE FUNCTION
  statement of the UDF, DB2 recognizes that some "continuity" is needed between
  successive invocations of the UDF, and therefore the implementing Java class
  is not instantiated for each call, but generally speaking once per UDF
  reference per statement. Generally it is instantiated before the first call
  and used thereafter, but may for table functions be instantiated more often--
  see the NO FINAL CALL execution model in the subsection which follows this
  one.
 
  If, however, NO SCRATCHPAD is specified for a UDF, either a scalar or table
  function, then a clean instance is instantiated for each call to the UDF, by
  means of calling the class constructor.
 
  A scratchpad may be useful
  for saving information across calls to a UDF.
  Whereas for C, C++ and OLE UDFs
  the scratchpad is maintained by DB2, in Java the UDF can simply use instance
  variables. For table functions using a scratchpad, you can control when you
  get a new instance by use of the FINAL CALL / NO FINAL CALL option on the
  CREATE FUNCTION statement, as indicated by the execution models in the
  subsection which follows this one.
 
  For scalar functions, you use the same
  instance for the entire statement.
  Another choice you have is to use the DB2
  provided Java methods getScratchpad() and setScratchpad(), in which case you
  will be using the DB2-managed scratchpad. The ability to achieve continuity
  between calls to a UDF by means of a scratchpad, is controlled by the SCRATCHPAD
  / NO SCRATCHPAD option of CREATE FUNCTION, regardless of whether the DB2 scratchpad or
  instance variables are used.
 
  Please note that every
  reference to a Java UDF in a query is treated independently, even if the same
  UDF is referenced multiple times. This is the same as what happens for OLE, C
  and C++ UDFs as well. At the end of a query, if you specify the FINAL CALL
  option for a scalar function then the object's close() method is called. For
  table functions the close() method will always be invoked as indicated in the
  subsection which follows this one.  If you do not define a close() method for
  your UDF class, then a stub function takes over and the event is ignored.
 
  And in the next paragraph, which begins "If you specify the ALLOW PARALLEL
  ...", the last sentence should be removed.
 
  Following the subsection "Changing How a Java UDF Runs", add a new subsection
  at the same level, entitled "Table Function Execution Model for Java", which
  contains the following:
 
  For table functions written in Java, it is important to understand what
  happens at each point in DB2's processing of a given statement which is
  significant to the table function. The table which follows details this
  information. The bottom part of each box hints what the code might be written
  to do for a typical table function which pulls some information in from the
  Web). Covered are both the NO FINAL CALL and the FINAL CALL cases, assuming
  SCRATCHPAD in both cases.
 
                  |                              |
      Point     | |    NO FINAL CALL             |    FINAL CALL
      in        | |    LANGUAGE JAVA             |    LANGUAGE JAVA
      scan      | |    SCRATCHPAD                |    SCRATCHPAD
      time      V |                              |
      ------------+------------------------------+------------------------------
      Before the  | No calls.                    | Class constructor is called
      first OPEN  |                              |  (means new scratchpad).
      for the     |                              | UDF Method is called with
      table func. |                              |  "FIRST" call.
                  |                              |- - - - - - - - - - - - - - -
                  |                              | Constructor initializes
                  |                              |  class and scratchpad
                  |                              |  variables.
                  |                              | Method connects to the Web
                  |                              |  Server.
                  |                              |
      ------------+------------------------------+------------------------------
      At each     | Class constructor is called  | UDF Method is called with
      OPEN of the |  (means new scratchpad).     |  "OPEN" call.
      table func. | UDF Method is called with    |
                  |  "OPEN" call.                |
                  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
                  | Constructor initializes      | Method opens the scan for
                  |  class and scratchpad        |  whatever Web data it wants.
                  |  variables.                  |  (Might be able to avoid
                  | Method connects to the Web   |  reopen after a CLOSE repo-
                  |  Server, and opens the scan  |  sition, depending on what
                  |  for Web data.               |  is saved in scratchpad!)
                  |                              |
      ------------+------------------------------+------------------------------
      At each     | UDF method is called with    | UDF Method is called with
      FETCH for   |  "FETCH" call.               |  "FETCH" call.
      new row of  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
      table func  | Method fetches and returns   | Method fetches and returns
      data.       |  next row of data, or EOT.   |  next row of data, or EOT.
                  |                              |
      ------------+------------------------------+------------------------------
      At each     | UDF method is called with    | UDF Method is called with
      CLOSE of the|  "CLOSE" call.               |  "CLOSE" call.
      table func. | "close" method is called if  |
                  |  it exists for class.        |
                  |- - - - - - - - - - - - - - - |- - - - - - - - - - - - - - -
                  | Method closes its Web scan   | Method might reposition to
                  |  and disconnects from the    |  the top of the scan, or
                  |  Web Server.                 |  close the scan.  It can
                  | "close" does not need to do  |  save any state in the
                  |  anything.                   |  scratchpad, which will
                  |                              |  persist.
                  |                              |
      ------------+------------------------------+------------------------------
      After the   | No calls.                    | UDF Method is called with
      last CLOSE  |                              |  "FINAL" call.
      of the table|                              | "close" method is called if
      func.       |                              |  it exists for class.
                  |                              |- - - - - - - - - - - - - - -
                  |                              | Method disconnects from the
                  |                              |  Web Server.
                  |                              | "close" does not need to do
                  |                              |  anything.
                  |                              |
      ------------+------------------------------+------------------------------
 
  Some notes on this table:
 
  1.  By "UDF Method" we mean the Java class method which implements the UDF-
      the method identified in the EXTERNAL NAME clause of the CREATE FUNCTION
      statement.
 
  2.  For table functions with NO SCRATCHPAD specified, the calls to the UDF
      method are as indicated in this table, but because the user is not asking
      for any continuity via a scratchpad, DB2 will cause a new object to be
      instantiated before each call, by calling the class constructor. It is
      not clear that table functions with NO SCRATCHPAD (and thus no
      continuity) can do very useful things, but they are supported.
 
  3.  These execution models imply the existence of a new class method which
      which enables the UDF to identify the call type. This method is provided
      with DB2 in the COM.ibm.db2.app.UDF class, and will be defined as:
 
             public int getCallType () throws Exception
 
  4.  These models are TOTALLY COMPATIBLE with what happens with the other UDF
      languages: C and C++ and OLE.
 
  In Chapter 15 of the "Embedded SQL Programming Guide", under the heading
  "Classes for Java Stored Procedures and UDFs", and further under the
  subheading "COM.ibm.db2.app.UDF", the paragraph describing the "public void
  close()" method should be changed to read as follows:
 
  This method is called by the database for scalar functions at the end of a
  statement evaluation, if the UDF was created with the FINAL CALL option, and
  is analogous to the final call for a C function. For table functions, it is
  called after the CLOSE call to the UDF method (if NO FINAL CALL is coded or
  defaulted), or after the FINAL call (if FINAL CALL is coded). If a Java UDF
  class does not implement this function, a no-op stub function will handle and
  effectively ignore this event.
 
  Immediately following the description for "public void close()", add material
  for the new method "public int getCallType() throws Exception", which reads
  as follows:
 
  This function is used by table function UDF methods to find out the call type
  for a particular call. It returns a value as follows (symbolic defines are
  provided for these values in the Com.ibm.db2.app.UDF class definition):
 
  o   -2 FIRST call
 
  o   -1 OPEN call
 
  o   0 FETCH call
 
  o   1 CLOSE call
 
  o   2 FINAL call
 
 
| 8.11.2  EXTERNAL UDF ERROR PROCESSING MODEL FOR SCALAR FUNCTIONS
 
| In chapter 7, in the section entitled "Other Coding Considerations", and the
| subsection entitled "Hints and Tips", there should be a new item 15 added to
| the list as follows:
 
| The error processing model for scalar UDFs which are defined with the FINAL
| CALL specification is as follows:
 
| 1.  If FIRST call fails, no further calls are made.
 
| 2.  If NORMAL call fails, no further NORMAL calls are made, but the FINAL
|     call is made.
 
| This means that if an error is returned on a FIRST call, the UDF must clean
| up before returning, because no FINAL call will be made.
 
| If FINAL CALL is not specified for the scalar UDF, then only NORMAL calls are
| made, and they cease if an error is returned.
 
| There should also be a new item 16 added to the list as follows:
 
| The error processing model for table functions is defined in the "Table
| Functions Considerations" section of this chapter.
 
 
| 8.12  NOT ATOMIC COMPOUND SQL
 
| DB2-Connect does not support SELECT statements selecting LOB columns in a
| compound SQL block.
 
 
  8.13  OTHER REVISIONS TO THE BOOK
 
  The following changes related to the Embedded SQL Programming Guide have been
  documented in the Release Notes for FixPaks prior to this release.
 
 
  8.13.1  REVISIONS TO CHAPTER 1
 
  In chapter 1 of the ESPG, in the first paragraph
  under the heading "DB2 Bind File Dump Tool-db2bfd", there is a reference to
  "Version 3"; this should be changed to Version 5.
 
 
  8.13.2  REVISIONS TO CHAPTER 4
 
  In chapter 4, the section "Join Strategies in a
  Partitioned Database" has several subsections showing examples of the
  different join strategies. In the pieces of artwork, all occurrences of
  "LINEITEMS" should be changed to "LINEITEM". Also, the subsection "Broadcast
  Inner-Table Joins", has incorrect text in the example shown. The text of the
  example should be:
 
       The LINEITEM table is sent to all database partitions that have the
       ORDERS table. Table queue q3 is broadcast to all database partitions
       of the outer table.
 
 
  8.13.3  REVISIONS TO CHAPTER 7
 
  The following changes pertain to chapter 7:
 
  o   Section OLE Automation UDFs in C++, at the end of the section, in the
      code:
 
           short indicator1,  // input indicator
           short indicator2,  // output indicator
 
      should be changed to
 
           short *indicator1,  // input indicator
           short *indicator2,  // output indicator
 
  o   Section Hints and Tips, item number 4: "reason code 21" should be changed
      to "reason code 34".
 
  o   The following changes pertain to the section entitled "The Arguments
      Passed from DB2 to a UDF":
 
      -   For the SQL-state argument, for the sqlstate 02001, it should read:
 
               02001    Only valid for the FETCH call to table functions, it means
                        there are no more rows in the table.
 
      -   For the dbinfo argument, list item "16. Table function column list
          (tfcolumn)", replace the first paragraph with the following:
 
               If this is a table function, this field is a pointer to an array
               of short integers which is dynamically allocated by DB2.  If
               this is a scalar function, this pointer is null.
 
  o   The following changes pertain to the section entitled "Using LOB Locators
      as Parameters or Results":
 
      -   For the figure showing (function header) declaration of the APIs,
          entitled "DB2 Lob Locator APIs defined in sqludf.h", the descriptions
          of the API parameters have been improved. See the actual sqludf.h
          file for details.
 
      -   There are five APIs described in the section. Where the return codes
          from these APIs are described, "other" is cited as a possibility for
          each API. For each API, the description of "other" should be changed
          to the following:
 
 
               other   Invalid locator or other error (for example, memory error).  The value
                       that is returned for these cases is the SQLCODE corresponding
                       to the error condition.  For example, -423 means invalid
                       locator.  Please note that before returning to the UDF with
                       one of these "other" codes, DB2 makes a judgement as to the
                       severity of the error.  For severe errors, DB2 remembers that
                       the error occurred, and when the UDF returns to DB2, regardless
                       of whether the UDF returns an error SQLSTATE to DB2, DB2
                       takes action appropriate for the error condition.  For
                       non-severe errors, DB2 forgets that the error has occurred,
                       and leaves it up to the UDF to decide whether it can take
                       corrective action, or return an error SQLSTATE to DB2.
 
      -   Three of the five APIs describe a possible return code of -1, meaning
          that the locator passed on the call has already been freed. These
          APIs do not return a -1 return code to the UDF. This condition is
          handled under the "other" return code using SQLCODE -423 (invalid
          locator).
 
      -   In item 1: sqludf_length(), an additional return code value is
          defined:
 
               -3      Bad input value is provided to the API.  This includes:
                       - udfloc_p (address of locator) is zero
                       - return_len_p (address of where to put length) is zero
 
      -   In item 2: sqludf_substr(), an additional return code value is
          defined:
 
               -3      Bad input value is provided to the API.  This includes:
                       - udfloc_p (address of locator) is zero
                       - start is less than 1
                       - length is negative
                       - buffer_p (buffer address) is zero
                       - return_len_p (address of where to put length) is zero
 
      -   In item 3: sqludf_append(), an additional return code value is
          defined:
 
               -3      Bad input value is provided to the API.  This includes:
                       - udfloc_p (address of locator) is zero
                       - length is negative
                       - buffer_p (buffer address) is zero
 
      -   In item 4: sqludf_create_locator(), an additional return code value
          is defined:
 
               -3      Bad input value is provided to the API.  This includes:
                       - udfloc_p (address of locator) is zero
                       - loc_type is not one of the three valid values
                       - loc_p (address of where to put locator) is zero.
 
      -   In item 5: sqludf_free_locator(), an additional return code value is
          defined:
 
               -3      Bad input value is provided to the API.  This includes:
                       - udfloc_p (address of locator) is zero
 
      -   In item 3: sqludf_append(), add the following paragraphs before the
          discussion of return codes:
 
               This API can be used to build very large LOB values in a piecemeal
               manner.  In cases where a large number of appends is used to build
               a result, the performance of this task can be improved by:
               - allocating a large application control heap (APP_CTL_HEAP_SZ is
                 the database manager configuration parameter)
               - doing fewer appends of larger buffers; for example, instead of doing
                 20 appends of 50 bytes each, doing a single 1000 byte append.
 
               SQL applications which build many large LOB values via the
               sqludf_append() API may encounter errors caused by limitations on
               the amount of disk space available.  The chance of these errors
               happening can be reduced by:
               - using larger buffers for the individual appends
               - doing frequent COMMITs between statements
               - in cases where each row of a SELECT statement is building a
                 LOB value via this API, using a CURSOR WITH HOLD and doing COMMITs between rows.
 
      -   In item 5: sqludf_free_locator(), The paragraph beginning
          "Typically..." is incorrect. Replace it with:
 
               Use this API to free any locators that were created with the
               sqludf_create_locator() API, and which were used only for internal
               manipulation.  It is NOT NECESSARY to free locators passed into
               the UDF.  It is NOT NECESSARY to free any locator created by the UDF
               via sqludf_create_locator() if that locator is passed out of the
               UDF as an output.
 
      -   After the enumerated coverage of the individual APIs, add the
          following text:
 
               The following notes apply to the use of these APIs:
               1. A UDF which is defined to return a LOB locator has several
                  possibilities available to it.  It can return:
                  a. an input locator passed to it
                  b. an input locator passed to it which has been appended to
                     via sqludf_append()
                  c. a locator created to via sqludf_create_locator(), and appended
                     to via sqludf_append().
               2. A table function can be defined as returning one or more
                  LOB locators.  Each of them can be any of the possibilities
                  discussed in the preceding item.  It is also valid for such
                  a table function to return the same locator as an output for
                  several table function columns.
               3. A LOB locator passed to a table function as an input argument
                  remains alive for the entire duration of the row generation
                  process.  In fact, the table function can append to a LOB using
                  such a LOB locator while generating one row, and see the
                  appended bytes on a subsequent row.
               4. The internal control mechanisms used to represent a LOB which
                  originated in DB2 as a LOB locator output from a UDF (table
                  or scalar function), take 1950 bytes.  For this reason, and
                  because there are limitations on the size of a row which is
                  input to a sort, a query which attempts to sort multiple
                  such LOBs which originated as UDF LOB locators will be
                  limited to (at most) two such values per row, depending on
                  the sizes of the other columns involved.  The same
                  limitation applies to rows being inserted into a table.
 
 
  8.13.4  REVISIONS TO CHAPTER 9
 
  In chapter 9, in the section "Accessing DRDA
  Servers", the second paragraph in point 1 should read:
 
       For IBM products, consult the IBM SQL Reference, Version 5.2 (S10J-8165), before
       you start coding.
 
 
  8.13.5  REVISIONS TO CHAPTER 15
 
  The following changes pertain to chapter 15:
 
  o   The book states that the user's Java classes must implement the
      interfaces that are provided. However, since the Java library provided in
      "sqllib/java/db2java.zip" contains classes, the user classes must extend
      Java classes of the same name. References to
 
           class ... implements COM.ibm.db2.app.UDF
           { ... }
 
      should be changed to
 
           class ... extends COM.ibm.db2.app.UDF
           { ... }
 
      The same is true for "COM.ibm.db2.app.StoredProc".
 
  o   The book states that new LOB values can be created by UDFs and stored
      procedures by using the following classes/routines:
 
           COM.ibm.db2.app.Blob    public static Blob new() throws Exception;
           COM.ibm.db2.app.Clob    public static Clob new() throws Exception;
 
      These were renamed and moved to the new class COM.ibm.db2.app.Lob as the
      following functions:
 
           COM.ibm.db2.app.Lob    public static Blob newBlob() throws Exception;
           COM.ibm.db2.app.Lob    public static Clob newClob() throws Exception;
 
  o   The Java type mapping table includes the following mappings:
 
           SQL Type           Java Type (UDF)   Java Type (Stored Procedure)
 
           LONG VARCHAR       Clob              Clob
           LONG VARGRAPHIC    Clob              Clob
 
      These should be changed to:
 
           LONG VARCHAR       String            String
           LONG VARGRAPHIC    String            String
 
  The following information should be added after the second paragraph in the
  section "Java UDFs and Stored Procedures":
 
       When creating a stored procedure in the Java
       language, you must use the CREATE PROCEDURE
       statement to register the procedure to the system
       catalog table SYSCAT.PROCEDURES.  For more
           information, refer to the section on the CREATE
           PROCEDURE statement in the SQL Reference.
 
  The following is a limitation of the Java UDF/stored procedure support: On a
  mixed code page database server, Java user-defined functions and stored
  procedures cannot use CLOB type arguments, because random access on character
  boundaries on large mixed code page strings has not yet been implemented.
  Full support for all LOB types is intended for SBCS databases. For mixed
  databases, support is intended for the BLOB and the DBCLOB types. As a
  workaround, applications running on a mixed database system should convert
  CLOB arguments to DBCLOB, LONG VARGRAPH, or LONG VARCHAR types.  For UDFs,
  this can be done with the CAST operator.
 
  In appendix B, which describes the sample programs, references (in table 24)
  to the loblocud, db2uext2, and autoloader samples should be removed. For
  additional information, also refer to 8.13.6, "Revisions to Sample Programs"
  and 8.13.7, "Other File Changes".
 
  In the index, java_heap_sz is incorrectly referred to as java_heap_size.
 
 
  8.13.6  REVISIONS TO SAMPLE PROGRAMS
 
  Revisions to sample programs were made for
  various languages on the following platforms: AIX, HP-UX, OS/2, Solaris,
  Windows NT, and Windows 95. The revisions are to correct some code errors and
  to make better use of API calls within the programs. Some programs have been
  made generic for all platforms:
  before they were only available on specific
  platforms; for example, 'makeapi.sqc'. Also, the interface has been improved
  for several programs to better guide the user on program input, and to give
  more informative output.
  Many redundant files have had their file contents
  deleted, replaced by the comment "This file is not needed". Where a new file
  replaces the redundant file's functionality, a reference to this is also
  provided.
 
  The following is a list of the updated files. Each entry contains the name of
  the file, its language, and the platforms for which it has been changed.
  Where the same file has been revised in more than one language, one file name
  is given with multiple extensions, as in "dbconf.c/.cbl/.f/.for". "COBOL"
  refers to both IBM and Micro Focus COBOL, unless otherwise indicated. "All
  platforms" refers to the following platforms only: AIX, HP-UX, OS/2, Solaris,
  Windows NT, and Windows 95. In some situations, a file may not be available
  on a specific platform; if you are not sure, refer to the README file in the
  appropriate samples directory. The file entry also contains a comment
  describing the nature of the change. The changes are as follows:
 
  adhoc.sqc
            C for all platforms
 
            Fixed memory leak and improved user interface.
 
  blobfile.cmd
            REXX on AIX, OS/2, and Windows NT
 
            The correct file version is now available for OS/2.
 
  dbconf.c/.cbl/.f/.for
            C, COBOL, FORTRAN for all platforms.
 
            Fixed 'version' parameter error in API call. Improved the user
            interface. Recoded Micro Focus COBOL version to make it generic for all
            platforms.
 
  d_dbconf.cbl
            Micro Focus COBOL for all platforms.
 
            Recoded to make generic for all platforms.
 
  d_dbmcon.cbl
            Micro Focus COBOL for all platforms.
 
            Recoded to make generic for all platforms.
 
  da_manip.h/.sqc
            C for all platforms
 
            Redundant files: contents deleted.
 
  dbcat.f/.for
            FORTRAN for all platforms
 
            Fixed 'version' parameter error in API call.
 
  dbmconf.cbl/.cmd
            Micro Focus COBOL for all platforms and REXX on AIX, Windows NT,
            and OS/2
 
            Recoded Micro Focus COBOL version to make it generic for all
            platforms. The correct REXX version is now available for OS/2.
 
  dbinst.for
            FORTRAN for OS/2
 
            Improved user interface.
 
  dbsnap.cbl
            IBM COBOL for all platforms
 
            Fixed 'version' parameter error in API call.
 
  dbstat.sqb
            Micro Focus COBOL for all platforms
 
            Fixed 'version' parameter error in API call.
 
  db_udcs.for
            FORTRAN for OS/2
 
            This program is now available for OS/2 FORTRAN.
 
  ebcdicdb.for
            FORTRAN for OS/2
 
            This program is now available for OS/2 FORTRAN.
 
  fillcli.sqc
            C for all platforms
 
            This is a client program; the server program is fillsrv.sqc. These
            programs have been extensively rewritten, combining the
            functionality of the supporting programs da_manip, system, and
            tabinfo (which are now redundant). The use of APIs, demonstration
            of SQLDA structures, and user interface have all been improved.
 
  fillsrv.sqc
            C for all platforms
 
            This is a server program; the client program is fillcli.sqc (see
            that entry for more information).
 
  inpsrv.def
            Micro Focus on OS/2
 
            Redundant file: contents deleted, as compiler does not need
            definition file for stored procedures.
 
  inpcli.old
            C for all platforms
 
            Redundant file: contents deleted.
 
  inpsrv.old
            C for all platforms
 
            Redundant file: contents deleted.
 
  largevol.sqc
            C for all platforms
 
            This program is AIX-specific: file contents were deleted on the
            HP-UX, OS/2, Solaris, Windows NT, and Windows 95 platforms.
 
  lobeval.cmd
            REXX on AIX, Windows NT and OS/2
 
            The correct version is now available for OS/2.
 
  makeapi.sqc
            C for all platforms
 
            This replaces the specific program files for AIX and OS/2 with a
            generic program file that is now available for all platforms.
 
  monreset.f/.for
            FORTRAN program for all platforms
 
            Fixed 'version' parameter error in API call.
 
  monsz.cbl
            IBM COBOL for all platforms
 
            Fixed 'version' parameter error in API call.
 
  outsrv.def
            Micro Focus COBOL on OS/2.
 
            Redundant file: contents deleted, as compiler does not need
            definition file for stored procedures.
 
  inpcli.old
            C for all platforms
 
            Redundant file: contents deleted.
 
  inpsrv.old
            C for all platforms
 
            Redundant file: contents deleted.
 
  qload.sqb/.sqf/.sqc
            COBOL, FORTRAN, and C for all platforms
 
            Improved interface provides more information and eliminates
            extraneous 'load file' error.
 
  rechist.c/.sqc
            C for all platforms
 
            The non-embedded SQL program rechist.c has been replaced by the
            embedded SQL program rechist.sqc. The new file corrects a database
            connect error that occurred with rechist.c.
 
  sws.cbl
            IBM COBOL for all platforms
 
            Fixed 'version' parameter error in API call.
 
  system.c/.h
            C for all platforms
 
            Redundant files: contents deleted.
 
  tabinfo.c/.h
            C for all platforms
 
            Redundant files: contents deleted.
 
  tabspace.sqb/.sqf
            COBOL and FORTRAN for all platforms
 
            Improves use of APIs to now demonstrate multiple table spaces.
 
  tblsrv.c
            C for all platforms
 
            Corrected error in API call to shared library.
 
 
  8.13.7  OTHER FILE CHANGES
 
  The makefiles, where needed, have been updated to
  reflect the changes described in the previous section. For example, the C
  makefiles no longer create and link in the object files da_manip.obj,
  system.obj and tabinfo.obj for the stored procedure fillsrv.
 
  The OS/2 command files for IBM COBOL and FORTRAN, and their respective
  makefiles, now use a stack size of 64000: the stack size link option for
  COBOL is now "/ST:64000", and for FORTRAN it is now "stack=64000". It was
  found that some of the stored procedure programs needed a larger stack size
  to run correctly.
 
  The following command and batch file changes have been made:
 
  BLDF32.CMD
            FORTRAN for OS/2
 
            Redundant file: contents deleted.
 
  EMBPREP.BAT
            COBOL for Windows NT and Windows 95
 
            The correct version of this file is now available.
 
 
  9.0   DB2 ڷ׻
 
 
  9.1   CCA ڷ׻̯ DB2 Τ
 
  Ϊڷ׻ΤشĸЬ̯˷ӡ
  ˷˷شĸ
  ɢ

 
   SEARCH ڷ׻ TCP/IP ꢡ
  TCP/IP  DB2 Τ˷ѡ
  ڷ׻ĸȺ IP ߡڷ׻
  ĸ TCP/IP Τ (DNS) Ϊ DNSЬ
  ڷ׻ΪȴԶƦ
  DB2 Τ˷ĸ TCP/IP ȴΤĸ
  ꢡڷ׻ĸ DNS 
  ̯ IP ߡ
 
 
  10.0  عٶ
 
 
  10.1  DBI1768W 
 
   "ٯɢ" Ķ "sqllib/profile" ٶҳ
  "sqllib/db2profile"
 
 
  10.2  SQL0270N /̧
 
  ع SQL0270N Ķա
 
     "" Ķ
 
       23  REPLICATED Ʒ֪
 
     "ٯɢ" Ķ
 
       23  ƪ REPLICATED  CREATE TABLE 
           ֪㡤
 
  Ķ̥Զع SQL0270N
 
     "" Ķ
 
       30  ŧ֪ܩ CREATE SCHEMA
           Ȣ
 
       31   500 Զĸš޴
 
     "ٯɢ" Ķ
 
       30   CREATE SCHEMA Ȣŧܩ
           ŧ֪ CREATE Ȣ

 
       31  ŷš޴
 
 
  10.3  SQL0351N 
 
  عɡ select-list  "<position-number>"  LOB 
        DATALINK 衢 DRDA ع̶ƪ
 
  ޡLOB  DATALINK 衢 LOB  DATALINK ҳ
        ֱΡ DRDA ع̶ӡ"<position-number>"
         select-list 㡢 LOB  DATALINK 

 
  ȺȢݨܡ
 
  ٯɢ DRDA  LOB  DATALINK 롢ڷ
  input-list ̥衤 LOB ơ CHAR
  (SUBSTR(CLOB)) ̽ԫ DATALINK ơ
   DURLCOMPLETE ̪
 
       sqlcode:  -351
       sqlstate:  56084
 
 
  10.4  SQL0352N 
 
  عɡ input-list  "<position-number>"  LOB 
  DATALINK 衢 DRDA ع̶ƪ
 
  ޡLOB  DATALINK  DRDA ع̶ӡ
  "<position-number>"  input-list 㡢 DATALINK 
  
 
  ȺȢݨܡ
 
  ٯɢ DRDA  LOB  DATALINK 롢ڷ
  input-list 衤ƶġ DLVALUE ̪
  ԫ
 
       sqlcode:  -352
       sqlstate:  56084
 
 
  10.5  SQL0357N 
 
  عɡDB2 File Manager "<server-name>" Щ
  
        = "<reason-code>"
 
  ޡ̽ DB2 File Manager "<server-name>" 롢
  DATALINK ԫٶDB2 File Manager Щġ
  
 
       01 DATALINK ԫ DB2 File Manager 
 
       02 ɢջΤ¡ջ DATALINK
          ԫ DB2 File Manager ش
 
       03  DB2 File Manager ĸٯߦɢ㡤
 
       04 DATALINK ԫ DB2 ܡȢջش
          ġԶϡΤ¡
 
       05   DATALINK ԫ DB2 File Manager 먡

 
  ȺȢݨܡ
 
  ٯɢٯɢĶ
 
       01 DB2 File Manager ƫˡѥܡ
          ˷ܡϯҥ DB2 File ManagerΪ㡢
          ЬŢȢջѥܡ
 
       02  DB2 File Manager شջΤ¡ջ
 
       03 Է DB2 File Manager ĸٯߦɢȩ
          Ϊ㡢ȩĸ 02 ů
          شƫٯߦɢɳȩġ

 
       04 ջ DB2 File Manager ĸ

 
       05  DB2 File Manager ٯ
          Ϊ㡢
          ЬŢȢջѥܡ
 
       sqlcode:  -357
       sqlstate:  57050
 
 
  10.6  SQL0432N 
 
  SQL0432N ҳٶ먡

 
  عɡٶشƫȴϯٶ
  "<udt-name>"
 
  ޡȢٶشҳ̦ȴϯ "<udt-name>"
  ҳ̦ȴ "<udt-name>" ٶΡĸĶ
  ٶشƫϯٶɢҳ̧Ρ
  ġǹ (INSERT  VALUES ƻ UPDATE  SET
  ƻ) CAST ݽ֪ڰ̱ȩϯֱ
  ٶΡȺȢݨܡ
 
  ٯɢҳٶشڰ̱ȩϯֱΡ
  ٶΡƶġڰ̱ϯֱ
  ȩҳ̧Ρڰ̱ٶ
  ȩҳ̧Ρ

 
       sqlcode:  -432
       sqlstate:  42841
 
  SQLSTATE 42841 ҳ
 
       42841  ٶشϯٶΡ
 
 
  10.7  SQL0680N 
 
  ԺޡġԶ֪թ޲֪Ϩ
  ֪̱Ķ
 
  o   4KB թ 500 Զ
 
  o   8KB թ 1012 Զ
 
  ƫ֪թ޲֪ȴƫݨܡ
  Ȣ֪Ϩ桤

 
  ٯɢΪݨܡ먡ЬǡġԶ޲
  ֪Ϩ (Ϊǡ)ΪȺ֪Ϩ㡢
  Ьڷּ֪̽ġԶԶ

 
 
  10.8  SQL0903N ̧
 
  ع SQL0903N  3 ҳ
 
       3. ٶȺɢ̧ġԶ DB2 File Manager ެ COMMIT

 
 
  10.9  SQL0911N ̧
 
  message SQL0911N ȴġԶ
 
       72. ܬٯߦϡҳܬٯ DB2 File Manager 
           먡
 
 
  10.10  SQL1339N 
 
  Ķع "" 毡
 
        CLI/ODBC Ȣش
       CLI/ODBC Ȣƫ SQLGetDiagRecSQLGetDiagField
        SQLError API̽ڵġԶع
 
 
  10.11  ܢ SQL5048N  SQL0901N
 
  Ϊ DB2  5.2 ܢع SQL5048N 
  SQL0901NȺ޲ƫҡջΤȹ
  ջڷ׻ȹ

 
  عƫ CONNECT  ATTACH ߡȺ
   AUTHENTICATION=CLIENTΪЬ
  ջɱջ AUTHENTICATION=SERVER
 
  Ϊĸ CONNECT  ATTACH ߡȺ DB2
  Common Server  2  DDCS  2 ϯ
  Ьջɱջ AUTHENTICATION=CLIENT
 
 
| 10.12  ع SQL5156NDBM ٶ TRUST_ALLCLNTS ԫ
 
| SQL5156N ջܡȢٶ "TRUST_ALLCLNTS" ԫ NO
| YES  DRDAONLY ̧ġԶ
 
| ޡٶ "TRUST_ALLCLNTS" ԫĶ
 
| 1.  NO = 0
 
| 2.  YES = 1
 
| 3.  DRDAONLY = 2
 
| ȴǶա
 
| ٯɢ롢 "TRUST_ALLCLNTS" ȴԫ
 
 
  10.13  SQL20053N ̧
 
  Ķع SQL20053N
 
     "" Ķ
 
       6.  ŧܩ㡢ġԶܩ FROM
           ƻ֪ܩС OUTER (Ϊܩ
            OUTER)
 
     "ٯɢ" Ķ
 
       6.  ΪġԶ OUTER ܩСЬ FROM ƻ
           ȩ OUTERΪܩ OUTER OUTER
           īܩ FROM ƻ
 
 
  10.14  SQL20056N ع
 
  Ķ SQL20056N ѥع
 
       DB2 File Manager "<server-name>" ĸݨܡܢ먡
        = "<reason-code>"̧ع "<diag-token>"
 
  ެ⡡  Щ "<diag-token>" ġѣҳ "UNKNOWN"
 
 
  10.15  SQL30083N ̧
 
  Ķ (Ϊҳϯ
  )ع SQL30083N
 
     "" Ķ
 
       24.  ϯů/桤
            ҳϯ
            ߡ

 
 
  11.0  ʸī
 
 
  11.1   WINDOWS 95  NEC PC98 ĸ DB2
 
   Windows 95  NEC PC98 ĸǵħ DB2 롢
  ɢȩ 96% ϡȩҡ
 
 
  1.   IMQCRINS.EXE ݨܡݨܡѺ <CRTL><ALT><DELETE>
      ɢܡȢס̽ IMQCRINS ݨܡѺġĶ֡
      Ѻڡ
 
  2.  ȩ롢ٯ˷ӡ
 
 
  11.2   WINDOWS 95  NEC PC98 ĸ DB2
 
  Ϊ Windows 95  NEC PC98 ĸǵħ DB2 
  ɢȩЩϡȩɢơ

 
  1.   UNINSTNQ.EXE ݨܡݨܡѺ <CRTL><ALT><DELETE>
      ɢܡȢס̽ UNINSTNQ ݨܡѺġĶ֡
      Ѻڡ
 
  2.  ڷ DB2 ɴ imnnq_95 硤
 
 
  11.3  
 
 
  11.3.1  /Τܨ
 
  ҳ HP  NETWARE ܨ¡ IPX/SPX  HP-UX 
  11.0 ĸݷΡ
 
 
  11.3.2   OS/2 ڷ׻ڷ׻ܨ/
 
   OS/2 ڷ׻ DB2ȴ ADSTAR Distributed Storage Manager (ADSM)
   3  PTF (ȢԺ) 3
 
 
| 11.3.3  DBCS  OS/2 WARP SERVER V.4 ̱
 
| Ϊ DBCS  OS/2 Warp Server  4 ĸ DB2 UDB
| OS/2 S)ЬѺ̥Ķ̱ɻ˷ SYS3175
| ԡġȹѥΪѺ̥Ķ̱ɻ
| 
 
|  SYS3175 عܢġԶɡ OS/2 FixPak WX03005
| ۡ
 
 
  11.3.4  DB2 FOR AS/400  DB2 UNIVERSAL DATABASE FOR AIXOS/2
           WINDOWS NT  SNA 
 
  ڷ DB2 for AS/400 ̯ DB2 Universal Database for AIX 
  5.2 롢 SNA Ĩڷ DB2 for AS/400 
  ̯ DB2 Universal Database for OS/2 ů Windows NT  5.2 롢
  Ρ
 
 
| 11.3.5  ġԶĸTP  DB2 ġ
 
| ġԶĸTP  DB2 ġЬΪ
| Զ DB2 ԷġԶ TP ꢡЬ APPC ů TCPIP Ҧġ
| Զ DB2 ĨԶҦ DB2 
| 
 
 
  11.3.6  ǿ۾ǵ/DB2 FOR OS/390 ԺȢ
 
  ڷ DB2 for OS/390 ̽ DB2 Universal Database Τ롢
  ȴĶ DB2 for OS/390 PTF
 
  o   PTF UQ06843 for APAR PQ05771
 
  o   PTF UQ09146 for APAR PQ07537
 
 
  11.4  DB2 UNIVERSAL DATABASE ī ()
 
  ڷ DB2 ڷ׻ DB2 ΤĸİĶ
  DB2 Τ¡
 
  o   DB2 for OS/390 V5  V5 ĸ
  o   DB2 Universal Database for AIX V5.2
  o   DB2 Universal Database for OS/2 V5.2
  o   DB2 Universal Database for Windows NT V5.2
 
 
  11.5  DB2 UNIVERSAL DATABASE ī ( AIX ĸϯ)
 
  UNIX** ʸīܡ 15 ٶܡ̦(Nodelock
  Administration Tool, NAT) License Use Management
  Administration Runtime Kit (LUM ARK) ġǹ AIX ĸ
  ϯ˷ĸȴ NAT  LUM ARK
  LUM ARK 㡢NAT ƫعٶ LUM ǵ
  ƫݾ IBM License Use Management Web ׻
 
       http://www.software.ibm.com/is/lum
 
  Ķ޲ LUM ARK ¡ҳħġ LUM ARK ެǹ
  ī DB2 CD  /cdrom/LUM 㡤
 
 
  11.6   (ջš޴ٴΤ̯ DB2
  EXTENDED ENTERPRISE EDITION FOR WINDOWS NT/
  ϯ ID )
 
  DB2 Extended Enterprise Edition for Windows NT ʸīܡ
  ġϯǡջš޴ٴΤ
   DB2 ٯҳ Windows NT Τ٭Ȣ롢Ⱥī
  ΪЬϯ̦ެ
  ܡЬϯ̽

 
 
| 11.7  NT ĸ DB2START ȴع
 
|  "db2start"  ( "NET START" ) ٯ Windows NT
| ĸջܡȢ롢ΪǶ˷ٯЬ
| ǶˡϯġNT ǵ뢡 "DB2DIAG.LOG" 
|  "db2start" ̯Ƕ먡Զҡ
| Windows NT ĸջܡȢҳ NT Τ٭ȢΪ
| ɳٯΤ٭Ȣ㡢Ь먡
 
 
  11.8  ڷȹ
 
  ޲ŧ̦ڰ䡢ǶӼ袡
  ̽ġѥŷ
  ġӳ袡ƫ
  ̯ϯƫǡ٭ƫ

 
  ġ㡢Щȴ
   EXPLAINΪϡٶ
   (ȹĶ)ɷϯɱոܬ㡢
  ůߦ (Ϊ)

 
 
| 11.9  ڷ DB2 ȹ 5.0 Ū̯ȹ 5.2
 
| ȴ "db2upd52" Ķع̯ DB2  5.2 ߦ
| ֡ʸīܡĶ
 
|      DB2 Extended Enterprise Edition for UNIX ʸī
|      DB2 Personal Edition ʸī
|      DB2 UDB Enterprise Edition for OS/2 ʸī
|      DB2 UDB Enterprise Edition for Windows NT ʸī
 
| ֪
 
  +------------+--------------------------------------------------------------+
| |            | Ϊ DB2 ڷ 5 Ū̯ 5.2Ʒ         |
| |            | DB2UPD52 Ϊ DB2 ڷ 2 Ū̯ 5.2        |
| |            | Ьջů SYSFUN                    |
  +------------+--------------------------------------------------------------+
 
|  DB2  5.2 ơSYSIBM.SYSDATATYPES ֪
| "BIGINT"ů SYSFUN ̧̪롢
| Ūҳ 5.2  DB2 ȴġԶջ޲
| DB2UPD52  "BIGINT"  SYSFUN ̪
| ΡĶ
 
| o    "SYSIBM.BIGINT" ̯ SYSIBM.SYSDATATYPES ֪㡤
 
| o   Ķ̪̯ SYSIBM.SYSFUNCTIONS ֪㡨
 
|          create function sysfun.abs(bigint) returns bigint ...
|          create function sysfun.absval(bigint) returns bigint ...
|          create function sysfun.ceil(bigint) returns bigint ...
|          create function sysfun.ceiling(bigint) returns bigint ...
|          create function sysfun.floor(bigint) returns bigint ...
|          create function sysfun.mod(bigint,bigint) returns bigint ...
|          create function sysfun.power(bigint,bigint) returns bigint ...
|          create function sysfun.round(bigint,integer) returns bigint ...
|          create function sysfun.sign(bigint) returns bigint ...
|          create function sysfun.trunc(bigint,integer) returns bigint ...
|          create function sysfun.truncate(bigint,integer) returns bigint ...
 
| DB2UPD52 ľ "REAL" ̯ SYSIBM.SYSDATATYPES 
| ֪㡢 UCASE (SYSFUN.UCASE) ̪ (Ϊ) 
| ̯ SYSIBM.SYSFUNCTIONS ֪㡤ȴ DB2UPD52 Ρٶ
|  Command Reference̽ڵع
 
|  "db2upd52" عٶDB2 Universal Database
| for UNIX ʸī  5.2 6  "ڷ DB2 ȹ 5.0 Ū̯
| ȹ 5.2"
 
 
| 11.10  LOTUS DOMINO GO WEB Τ DB2 ٯȢ
 
| Ϊ Lotus Domino Go web Τ¡ЬĶ먡
 
|      java.sql.SQLException:  handle  DB2 ٯȢ먡rc=-1
 
| ٯ web Τ¡ db2instanceĶ
 
|      startsrc -e "DB2INSTANCE=yourInstanceName" -s httpd
 
| ľ "httpd.conf"  CLASSPATH ů LD_LIBRARY_PATH 
|  (Զɱ "JavaClassPath" ů "JavaLibPath")
| ԯ UDB  5.2 
 
 
  11.11  Ȣ ( JAVA Ȣ)
 
  DB2  Java ʩ Java ϯ̪ (UDF)  HP-UX
   SCO UnixWare Τĸ̽ DB2 ջ
 
 
  11.12  DB2 ܨթ ()
 
   DB2 for Solaris  DB2 for SCO UnixWare 7 Ż
  ѡڷ "pt" ҳ "pt_BR"
 
 
  12.0  DB2 File Manager ʸī
 
 
| 12.1  ڷ׻īů
 
| ȴ FixPak 7 ڷ׻ĸǡڷ׻Ʒ
| ľȴ FixPak 7 ڷ׻ĸī
 
 
| 12.2  DB2 DATA LINKS MANAGER  DB2 ENTERPRISE -
| EXTENDED EDITION ˷ĸ
 
| DB2 Data Links Manager (ġҳ "DB2 File Manager") Ʒ
| DB2 for AIX Enterprise Edition ġData Links Manager 
|  AIX ɢ˷ĸջġ DB2
| Enterprise - Extended Edition (EEE) for AIX ġDATALINK
| ԫ DB2 EEE ˷ӡ⡢DATALINK  EEE ˷
| ĸٶ DB2 Data Links Manager ĸ
 
 
  12.3  DLFM ܡϯ ID
 
  Ⱥȹ DLFM ܡϯ ID  "dlfm"
 
 
  12.4   NFS ̽ DLFS ˷
 
   "READ PERMISSION DATABASE"  DATALINK ĸ SQL
  SELECT ̦ȴ̽ش URL ءȺ̽
  شȴġԶ DL_EXPINT ջٶ NFS
  ڷ׻̽롢ƫܢ̽شȴ桤
  ҳ NFS ڷ׻ȩʸ̽ (Զ 12 Զ) ƫ
  Զ
 
   NFS ڷ DLFS ˷롢ľ
  "noac"
 
       mount -o noac birdcage:/dlink /dlink
 
 
  12.5   DATALINKS ڷ׻̱
 
  Datalinks ڷ׻Ρ
 
  o    Windows 3.1 ɢ˷ĸ
  o    DB2  5.2 ȹĶ (AIX ڷ׻)ľƫ
      FixPak U453782  5.0  5.0 ĸ
 
 
  12.6  DLFM 
 
  DLFM ҳ̿ dlfm_db Ϊ
  ЬԫȺƫ
  ΪЬԫĶ

 
       dlfm setup [<dlfm_db_config_file>]
 
  Ķҳ dlfm_db.cfg
 
    # Sample dlfm_db.cfg file for DLFM setup
    # Please note: the equal sign is needed between the option name and the value.
    # Any of the updateable DB2 database configuration parameters can be included in this file.
    # There are also three special DLFM-specific options which the user can specify:
    # DLFM_DB_DIR : directory where the dlfm_db database is to be created
    # DLFM_DB_CATTBL_DIR : directory where the catalog table space is to be located
    # DLFM_DB_TMPTBL_DIR : directory where the temporary table space is to be located
    # If there are no values for these options, or the options are not present,
    # default values are used.
    ####################
    DLFM_DB_DIR=
    DLFM_DB_CATTBL_DIR=
    DLFM_DB_TEMPTBL_DIR=
    LOCKTIMEOUT=60         # Default and recommended value for dlfm_db is 60
    LOGFILSIZ=2001         # Default and recommended value for dlfm_db is 2001
    LOGPRIMARY=6           # Default and recommended value for dlfm_db is 6
    MAXAPPLS=128           # Default and recommended value for dlfm_db is 128
    MAXLOCKS=100           # Default and recommended value for dlfm_db is 100
 
 
  12.7  ܡȢΤĸڷߦ
 
  ܡȢΤĸѥߦϡů
  ܡϯˤߦ٭ެǹܩ

 
  ެ⡡ DB2 File Manager ĸӼ㡢Ь DB2 File
        Manager ٯ DB2 Ȣƫࡤ FORCE
        APPLICATION ڰȢ˷ӡ DB2
        ΤĸӼ㡢Ьľ "dlfm shutdown"
        ̿ DB2 File Manager
         ѥԯȹ
        ϡ
 
  ΪϯڡЬ˷ġ
  ڡ]ڷ٭ߦЬܡϯɢ˷
  ެǹϯ鬡ߦ
  ߦ˷ȴ⡢ůشߦ
  ˷ѥſ˷ެǹѥա
  Ⱥѥ̿
  ȴ֪ RECONCILE ȴȢ
  "db2_reconcile_aid" ( sqllib\misc) ŧȺɢơ
  Ķ
 
  ެ⡡
 
  1.  ȴ RECOVERY NO ĶȢݨܡ
 
      o   Ϊ˷ĸ̯DATALINK ԫҳ NULL
      o   Ϊȴ̯̦ȴ READ  WRITE PERMISSION FS
          Ьġ硢̡

      o   Ϊȴ̯̦ȴ WRITE PERMISSION BLOCKEDЬ
          ԺѡΪԫ١DATALINK
          ԫҳ NULL
 
  2.  ȴ RECOVERY YES ĶȢݨܡ
 
      o   Ϊ̯ԺԺ桢
          ЬڷΤߦۡ

      o   ΪԺԺ桢Ь DATALINK
          ԫҳ NULL
 
  3.  ̡Τĸݨ硤
 
  ެǹůߦ˷
 
  Ķ AIX  4.1.5ƫ̧ɢ˷ӡϯ
   root īެǹߦɢơ

 
  ެǹ
 
   3 inode ֪Ȣɢެǹ
 
  1.  ެǹ˷ӡ
 
           ˷ܡ --> ˷ --> ˷
 
  2.  ެǹ˷ӡ
 
           ˷ܡ --> ˷ --> ެǹ˷
 
      Ȣެǹܩƫŷߦ桤
      ȹ 0 ެǹȹ 1 ̯ 9 Ȣެǹ
      ȹ n ެǹƷެǹڷȹ (n-1) ެǹѥ
      ȹ n ެǹѥĶġԶެǹȹ (n+1)
      ެǹ
 
      ľƫڷƵެǹ
 
           % backup -<level> -uf <backup device> <filesystem>
 
           
              ެǹ
                 % backup -0 -u -f/dev/rmt0 /dlfs1
              ĶġԶȢެǹ
                 % backup -1 -u -f/dev/rmt1 /dlfs1
              ȺѥȢެǹҳ
                 % backup -2 -u -f/dev/rmt2 /dlfs1
 
  3.  ެǹ˷ӡ
 
           ˷ܡ --> ˷ --> ˷
 
 
  ߦ
 
  ǿЩެǹġߦ
 
  1.  ߦ˷
 
           ˷ܡ --> ˷ --> ˷
 
  2.  ߦ˷ӡߦެǹ֪Ȣ

 
           ˷ܡ --> ˷ --> ߦ˷
 
 
  ľƫڷƵߦڷߦ (ǩɻ˷
  ) Ķ

 
       % restore -r -f <backup device>
 
  ΪȢެǹܩЬߦ޲ȹ 0 ެǹѥߦȹ
  0 ެǹѥȹ 1 ެǹ
  ߦȹ (n-1) ެǹѥȹ n ެǹ˷ root
   "restoresymtable" Ȣެǹߦѽȡ

 
  ߦŦ "-M" "-M" ش
  ȩҳߦ˷桤

 
  Ϊެǹߦɢ ADSM 硢Ьߦɢѥ
  inode ެǹǿЩެǹ

 
  ɢެǹա
 
  Ʒߦ
 
  ϯƫǨȢߦĸ "-i" ߦ

 
       % restore -i -f <backup device>
 
  ϯġǹެǹϯīǨ
  Ȣ Shellƫ "-h" ߦ硤

 
  ˷ܢ
 
  ش (err_log)/fsysadm.log 㡢̧ (err_log)
   /etc/sylog.conf 硤 Ⱥ
  ݷˡ  ġԶǵȴġԶ硤 ľش祡 ֪ȢĶ
 
       Time = <timestamp> EUID = <integer>  UID = <integer> GID = <integer> Mode = <octal>
       Action = <CREATE/REMOVE/SETATTR> Object type = <DIR/FILE> Path = <fully qualified name>
 
       ̧㡨
 
         Time Ҥٯ (ٴ)
         EUID ϯٯɢȴϯ ID
         UID ů GID ǡ/ϯ ID ů ID
         Ժ/ϯ ID ů ID
         Mode Ĭ/Ȣ
 
         Action ƫ CREATE (ǡ/)REMOVE (ƪ
         /)  SETATTR (ϯԺ/Ȣ)
         
         ǵƫ DIR ()  FILE ()
         Path 
         /硤
 
  ߦ˷ѥ̧ġӳ
 
  db2_reconcile_aid Ȣġ̱Τĸ
  ѥΤĸ DATALINK ƫġӳ
  ֪Ⱥ̱ů RECONCILEĶ

 
       db2_reconcile_aid -check -db <db_name> [-server_name <fileserver_name>]
          [-reportdir <report_directory>]
 
       ̧㡨
          -check:            ƫ̧ġӳ֪
                             ɢơ
          db_name:           ɢջꢡ

          fileserver_name:   ɢ DLFM ꢡΪ
                             ꢡȴΤ¡
          report_directory:  ġԶɢ硤
                             ġԶ֪ǡĶ֪Ȣ
                             <tbcshcema>.<tbname>.<ext>̧
                             <tbschema> ֪<tbname> 
                             ֪ꢡ <ext>  .ulk  .exp
                             .ulk ȴΤĸ
                              .exp ȴΤ
                             ĸعٶ
                             Command Reference  RECONCILE 
                             
 
 
| 12.8  ެǹ (DL_NUM_BACKUP)
 
| CONFIGURATION TYPE  ջ
 
| PARAMETER TYPE      ƫ
 
| DEFAULT [RANGE]     3 [ 1 - 31 ]
 
| ԶٶܡȢެǹع޲ެǹ (ջů
| ֪Ϩ) 
ڷջȺԫȭ̯㡤
| (1) ԫȩĶġԶ DB2 ެǹȭ̯㡤
 
| Ⱥٶԫ 1 Զެǹ
 
| Ⱥٶ "Recovery=Yes  DATALINK 
 
 
  12.9  DB2 File Manager ߦ٭
 
  Ƕ˷ӡƫߦȹ桤DB2 File Manager
  Զߦ٭ſġȹԶƫ
  ٭ůڷ٭ߦ
 
  
 
       DLFS ˷          - ش  (/dlink)
       DLFM ެǹ          - ެǹ (/home/dlfm/dlfm_backup)
       DLFM           - DLFM ϯ ID  (/home/dlfm)
       DLFM DB2 ջ        - ȴ meta  DB2 ջ (DLFM_DB)
       DB2 ջ             -  DATALINK شջ (CROWN)
 
  Щ
 
  ˷ĸ DLFM ջ (DLFM_DB)DLFS ˷ӡ
  DLFM ެǹ硣ů DLFM 硤ԯ

 
  ܡȢ ( IBM  ADSM) ĸެǹ DLFS ˷ DLFM
  ެǹ硤ƫǵ⡤
  
 
  ջġԶԶ DB2 File Manager ɭȴ̥Զջ
  ġԶ DB2 File Manager Ьߦ٭ڵ
  㡤

 
   DB2 ջջެǹ (ů DB2 File Manager ĸ
   DLFM_DB ջ)
 
  Ω桩
 
  ĶƾԶƫߦǵ
 
  o   ȴ DATALINK ֪ DB2 ջ
 
  o   DB2 File Manager ջ (DLFM_DB)
 
  o   DB2 File Manager  DLFS ˷
 
  o   DB2 File Manager ެǹ
 
  ĶšɱҳǵġԶħ DB2 File Manager ߦ٭
  

 
  Ķߦ٭ƫ DB2  DB2 File Manager
  ġɢơġԶЩ "DB2:"  "DLFM:" 
  ȩɢ

 
  ȴ㡢DB2 ջҳ "crown" DATALINK 
  DB2 ֪ҳ "datalinktable"
 
  SCENARIO #1A:
 
  桨
 
  DB2 ջ۹ƫ DB2 ެǹ
 
  ߦ
 
                         ɢ
 
       DB2:                  db2 "ߦջ crown"
       DB2:                  db2 "ROLL-FORWARD ջ crown ̯ʢ"
                                 (ߦ롢̯ʢ)
       DB2:                  db2 "֪ datalinktable ̧ġӳ"
 
  ٭ #1B:
 
  桨
 
  DB2 ջ۹ƫ DB2 ެǹ
 
  ߦ ۹ջɢȩջȴ
  ḡ
 
 
                         ɢ
 
       DLFM:                 db2 "̯ dlfm_db"
       DLFM:                 db2 "ڷ dfm_bdid ̽ dbiddbnamedbinsthostname"
       DLFM:                 db2 " dfm_dbid ٯɢ=5
                               ̧ dbid=x'35B3D7BE0006BF7B'"
                                 ( dbid Щġԫ)
       DB2:                  db2 "ߦջ crown"
       DB2:                  db2 "ROLL-FORWARD ջ crown ̯ʢ"
                                 (ߦ롢̯ʢ)
       DB2:                  db2 "֪ datalinktablȩġӳ"
                                 ()
 
  ٭ #2:
 
  桨
 
  DLFM_DB ¡ƫެǹ DLFM_DB ջȴ
 
  ߦ
 
                         ɢ
 
       DLFM:                 db2 "ߦջ dlfm_db"
       DLFM:                 db2 "ROLL-FORWARD ջ dlfm_db ̯ʢ"
 
  ٭ #3:
 
  桨
 
  DLFM_DB ¡ƫ DLFM_DB ջެǹȴƫ
 
  ߦ
 
                         ɢ
 
       DLFM:                 db2 "ߦջ dlfm_db"
       DB2:                  db2 "֪ datalinktable ̧ġӳ"
 
  ٭ #4:
 
  桨
 
  ƪ DLFS ˷ӡ
 
  ߦ
 
                         ɢ
 
       DLFM:                 ڷܡȢߦ DLFS ˷ӡ
       DB2:                  db2_recon_aid crown
 
  ٭ #5:
 
  桨
 
  ƪ DLFM ެǹ硤
 
  ߦ
 
                         ɢ
 
       DLFM:                 ڷܡȢߦ DLFM ެǹ硤
 
  ٭ #6:
 
  桨
 
  ƪ
  DLFS ˷ӡƪ DLFM ެǹ硤
 
  ߦ
 
                         ɢ
 
       DLFM:                 ڷܡȢߦ DLFM ެǹ硤
       DLFM:                 ڷܡȢߦ DLFS ˷ӡ
       DB2:                  db2_recon_aid crown
 
  ٭ #7:
 
  桨
 
  DLFM_DB ¡ƫެǹ DLFM_DB ջȴ ƪ
  DLFS ˷ӡƪ DLFM ެǹ硤
 
  ߦ
 
                         ɢ
 
       DLFM:                 db2 "ߦջ dlfm_db"
       DLFM:                 db2 "ROLL-FORWARD ջ dlfm_db ̯ʢ"
       DLFM:                 ڷܡȢߦ DLFM ެǹ硤
       DLFM:                 ڷܡȢߦ DLFS ˷ӡ
                                 (鬡)
       DB2:                  db2_recon_aid crown
 
  ٭ #8:
 
  桨
 
  DLFM_DB ¡ƫ DLFM_DB ջެǹȴƫ ƪ
  DLFS ˷ӡƪ DLFM ެǹ硤
 
  ߦ
 
                         ɢ
 
       DLFM:                 db2 "ߦջ dlfm_db"
       DLFM:                 ڷܡȢߦ DLFM ެǹ硤
       DLFM:                 ڷܡȢߦ DLFS ˷ӡ
                                 (鬡)
       DB2:                  db2_recon_aid crown
 
  ٭ #9:
 
  桨
 
  DB2 ջ¡DLFM_DB ¡ƫެǹ DLFM_DB ջ
  ȴƪ
  DLFS ˷ӡƪ DLFM ެǹ硤
 
  ߦ
 
                         ɢ
 
       DLFM:                 db2 "ߦջ dlfm_db"
       DLFM:                 db2 "ROLL-FORWARD ջ dlfm_db ̯ʢ"
       DLFM:                 ڷܡȢߦ DLFM ެǹ硤
       DLFM:                 ڷܡȢߦ DLFS ˷ӡ
       DB2:                  db2 "ߦջ crown"
       DB2:                  db2 "ROLL-FORWARD ջ crown ̯ʢ"
                                 (ߦ롢̯ʢ)
       DB2:                  db2 "֪ datalinktable ̧ġӳ"
 
  ެ⡡ROLL-FORWARD ̯  (PIT) ̦ȴȴ DATALINK 
  (ҳ "recovery = no") ֪ī DataLink_Reconcile_Pending
  (DRP) 衤 ȴ֪ ROLL-FORWARD ̯ PIT ѥī
  ġ DATALINK ԫѥ SET CONSTRAINTS Ȣ֪ī DRP 㡢
  Ȣ
 
  12.10  ڷݨܡ (CLP) ǡԺ
 
  ٶȺ
 
       db2 add file manager <hostname> <portnum>
 
  ݷٶȢ db2dlcfg ̽ ȢĶ
 
       db2dlcfg -d <database name> -a <action> [-s <server host name>] [-p <port number>] -h
 
       ̧㡨
         database name     ջ ()
         action            ٯɢ ()
                             LIST - ջȴ DB2 File Managers
                             ADD  - ջ̽ġ DB2 File Manager
         server host name   DLFM ܡȢꢡ
         port number        DLFM ܡȢ
         -h                عġȺ̧ݷܩ
                           멡
 
  
 
       db2dlcfg -d SAMPLE -a LIST
       db2dlcfg -d SAMPLE -a ADD -s myhost.mydomain.mycompany.com -p 10001
 
  Ӽڷĸġ DB2 File Manager 롢ĶȢ硨
 
  1.   db2stop
 
  2.  ƪȴ datalink.cfg  (ġǹެ)
 
  3.   db2dlcfg ȢΤΡǡ datalink.cfg  
      ȹʩΤ¡
 
  
 
       cat datalink.cfg
       1     myfs1.mydomain.mycompany.com 10001
       2     myfs2.mydomain.mycompany.com 10001
       4     myfs2.mydomain.mycompany.com 10001
       (ĵԶħ)
 
       mv datalink.cfg datalink.old
 
       db2dlcfg -d SAMPLE -a ADD -s myfs1.mydomain.mycompany.com -p 10001
       db2dlcfg -d SAMPLE -a ADD -s myfs2.mydomain.mycompany.com -p 10001
       db2dlcfg -d SAMPLE -a ADD -s nullfs.mydomain.mycompany.com -p 10001
       db2dlcfg -d SAMPLE -a ADD -s myfs4.mydomain.mycompany.com -p 10001
       (ĵԶϨԫ)
 
 
  13.0   5.2 Щشʸīܡ
 
 
  عĶ DB2  5.0 ʸīȴá
 
  o   DB2 UDB Personal Edition ʸī  5.0
 
  o   DB2 UDB Enterprise Edition for OS/2 ʸī  5.0
 
  o   DB2 UDB Enterprise Edition for Windows NT ʸī  5.0
 
  o   DB2 UDB Enterprise Edition - Extended for UNIX  5.0
 
 
  13.1  DB2 ܨ
 
   DB2 ȴ DB2 Application Developer's Kit ٶ
  ܨܢȴ̥ۡǩɻ IBM DB2 Personal Developer's Edition
  Version 5  IBM DB2 Universal Developer's Edition Version 5
  ܨ¡̥ OS/2Windows NTWindows 95
  Windows 98
 
 
  13.2  
 
  ǿ۾ǵԺȢȴá
 
 
  13.2.1  OS/2 
 
   "Τܨ" 㡢 DB2UDB Enterprise  Workgroup
  Edition Ķɢ˷عա

 
       ڷ                  ̯
       --------------------         ---------------------------------------------
       OS/2 Warp Server SMP         OS/2 Warp Server Advanced V4 SMP 
 
   "ڷ׻ܨ" 㡢DB2 Client Application Enabler
  for OS/2SDK Net.Data ľա
 
       ڷ                  ̯
       --------------------         ---------------------------------------------
       OS/2 Warp Server SMP         OS/2 Warp Server Advanced V4 SMP 
 
 
  13.2.1.1  ƫڷ׻̯Τ
 
   Windows 95Windows 98 OS/2 ڷ׻Τ
   IPX/SPX ꡤ  Novell ˶ѥʼԯ
  ɲ̧ IPX/SPX ع̶ȹ Windows 95 
  Windows 98  SPX I ȹů OS/2  SPX I ȹ
  Windows 95  Windows 98 ̶ٸ롢 SPX I ȹ١
   Windows 95  Windows 98 ĸ SPX II ȹ
  Ρ紡 OS/2 ̶ٸ SPX I 롢ѡ
 
  Novell Щ OS/2 SPX I Ρ
  SPX I ġӳƫɭš
  ϡΪϡȺƫ Novell 
  SPD147611 
 
 
  13.2.2  NetFinity Ȣ - WINDOWS NT
 
  Ū DB2 UDB 롢"Netfinity Ȣ" Τ٭Ȣ DB2
  UDB EEE for Windows NT ҡ
 
  ҳħ DB2 UDB EE Ū̯ DB2 UDB EEE DB2 UDB EEE
  Щ "Netfinity Ȣ" Τ٭Ȣ

 
 
  13.2.3  COMMUNICATIONS SERVER FOR WINDOWS NT - ԺȢ
 
   DB2 Connect Enterprise Edition for Windows NT  IBM
  Communications Server for Windows NT (CS/NT) 롢Ӽּ̽
   (SQL CANCEL) DB2 UDB  5.2 ( DB2 UDB
  Extended Enterprise Edition for Windows NT) ů APAR ԺȢ
  JR11529 ů JR11170 for CS/NT
 
 
  13.2.4  š޴ջϯ Client Application Enabler Ū
 
  Ӽ DB2 UDB V5.0 Client Application Enableṟ
   Extended Enterprise Edition (EEE) š޴ջܨ
  
 
   DB2 Connect  UDB  5 ̱ܡ
  š޴ջ˷ܡ䡤

 
  ȺŪ DB2 ڷ׻ UDB EEE ܡ䡤
  ġƫĶ

 
  o   ȹ̱̱

 
  o   ջެǹߦƫڷ硢ϯ
      ROLL-FORWARD ջ ROLL-FORWARD ˷먡
      ڷ׻ɢ׻ db2 ݨܡ񢡤
 
      ΪԶġȴԶЬ DB2  2 ڷ׻
      ެǹߦ
 
  ġơΪ̱ܡ DB2 UDB EEE Τ¡
  ЬŪ Client Application EnablerȺԺȢ޲
  ġԶ FixPak for DB2 UDB V5.0 
 
       AIX         U452196
       Solaris     U452197
       HP-UX       U452198
       OS/2        WR09003
       Windows NT  WR09004
       Windows 3.1 WR09006
 
  ԺȢƫڷĶ̽ڵ
  "http://www.software.ibm.com/data/db2/db2tech" ڷ׻ȹ
   ľƫڷڷ׻Ȣ̽ڵ DB2 UDB  DB2 Connect
   Enterprise Edition  5.2
 
  DB2 UDB FixPak Ȣԯ FixPaks 12345 6
  ů DB2 UDB EEE for Windows NT  for Solaris  5 
  Client Application Enabler ԺȢ
 
 
  13.2.5   Windows ODBC ϯ Client Application Enabler Ū
 
  ڷ Windows NT  Windows 95 ڷ׻ ODBC Ȣ롢ʼԯ
   DB2 Client Application Enabler ŪԺȢȹ
  WR09014 ĸ
 
  ȺԺȢƫڷĶ̽ڵ"http://www.software.ibm.com/data/db2/db2tech"
  ڷ׻ȹľƫڷڷ׻Ȣ̽ڵ DB2 UDB 
  DB2 Connect Enterprise Edition  5.2
 
 
  13.2.6  AIX ĸ ADSM ԺȢȹ
 
   AIX š޴ջ㡢 ADSM ĸ֪Ϩެǹѥ
  ΪߦջDB2 
 
  ̧ߦ DB2  ()عҳ
  "SIG11sqloEDUCodeTrap"Ⱥ ADSM ڷ׻Ժ
  Ȣ 2.1.0.6 ۡ
 
 
  13.2.7  CS/NT  Windows NT ĸ̥
 
  ܢ Windows NT ĸ DB2 Syncpoint Manager (SPM) 䡢
   IBM Communications Server V5.01 for Windows NT (CS/NT) 롢
  ȴ SNA ̥

 
  SNA-based  SPM Ķܨġǹ
 
  o   DB2 Connect Enterprise Edition for AIX
  o   DB2 Connect Enterprise Edition for OS/2
  o   DB2 Connect Enterprise Edition for Windows NT
  o   DB2 UDB Enterprise Edition for AIX
  o   DB2 UDB Enterprise Edition for OS/2
  o   DB2 UDB Enterprise Edition for Windows NT
  o   DB2 UDB Extended Enterprise Edition for AIX
  o   DB2 UDB Extended Enterprise Edition for Windows NT
 
  SPM  DB2 UDB Client Application Enabler  DB2 Connect EE
   SPM  SNA ̥̽衤

  ⡢ջȢƫ DB2 UDB Τĸ̽衤
 
 
  13.2.7.1  ش CS/NT ɢҳ Windows NT Τ٭Ȣ
 
   IBM Communications Server for Windows NT (CS/NT) ѥ
  شҳ Windows NT Τ٭ȢȺٯ롢ٯٯ
  CS/NT
 
  ش CS/NT ɢҳ Windows NT Τ٭ȢĶ
  Ƕ̧ġ
 
       csstart  -a
 
  ٯٯ CS/NTϯ
 
       csstart -a c:\ibmcs\private\your.acg
 
  ̧ C:\IBMCS\PRIVATE\YOUR.ACG  CS/NT ꢡ

 
  Ķȹ롢ٯٯ CS/NT

 
 
  13.2.8  WINDOWS NT ů WINDOWS 95 ĸ DCE 
 
  DB2 Universal Database  DB2 Connect  5.2 ĶΡ
 
  o   Windows NT ů Windows 95 ĸ Gradient DCE CDS
 
  o   Windows NT ĸ IBM DCE CDS
 
  ľħ̥Զ DLL
 
  o   db2dce.grd Gradient DCE CDS
 
  o   dcedce.ibm IBM DCE CDS
 
  ԫ"db2dce.grd" ҳ "db2dce.dll"
 
  ެ⡡DB2 Ʒī "db2dce.dll"Ϊ IBM DCE CDS 㡢
         "dcedce.ibm" ҳ "db2dce.dll"
 
  ħ APPCIPX/SPXNETBIOS ů TCP/IP ܢЩľ
  ħ Named Pipes (NPIPE) ع̶ DCE ΡĶ
  ̱ȢЩľƫ̦ȴ DB_Communication_Protocol
  ůĶԫ DCE ǵ
 
       NPIPE;Computer_name;Instance_name
 
  عٶ Administration Guide  G
 
 
| 13.2.9   WINDOWS NT ȹ
 
  Ӽ Windows NT ĸش̯ DB2 Ьȴ
  Microsoft Service Pack 3 for Windows NT 4.0
 
|  Microsoft Service Pack 4 㡢Windows NT ش
| شԺħ
 
 
  13.2.10  SYSADM  (UNIX)
 
   UNIX ʸī㡢ض̯롢
  ϯȴջܡȢ SYSADM_GROUP ٶΪ
   SYSADM_GROUP ٶЬ SYSADM ҳȴϯ
  ڡ

 
  ʸīٶȴϯġѣҳ̦ȴ SYSADM
  紡Ϊȴ SYSADM_GROUP ٶЬ
  ٶջܡȢ SYSADM ڡ

 
 
  13.3  ̧ɢ
 
  ɢá
 
 
  13.3.1   CS/NT ῡܡȢ(SPM)
 
  Ϊ SNA  SPMЬ IBM Communications Server for
  Windows NT  5.01 DB2  5.2ǩ TCP/IP  SPM
  "DB2 Connect Enterprise Edition ʸī" ġ SPM ҡ
 
  Ķ CS/NT  DB2 ConnectٶDB2
  Connect Enterprise Edition ʸīܡġ 14 
 
  1.  SPM  LUƫCS SNA / LU 6.2
       CS/NT
 
  2.  LU ƫĶȢȩȺ
 
      a.  ٯ CS/NT Ȣ
 
      b.  
 
      c.  Ⱥ
 
      d.   SPM  LU ̯Ķ

 
               LOCAL_LU=(LU_NAME=SPMNAME
                         LU_ALIAS=SPMNAME
                         LU_SESSION_LIMIT=0
                         NAU_ADDRESS=0
                         ROUTE_TO_CLIENT=0
                         SYNCPOINT_SUPPORT=0)
 
      e.   SYNCPOINT_SUPPORT=0 ҳ SYNCPOINT_SUPPORT=1
 
      f.  
 
  3.  鬡
 
  4.   DB2
 
  5.   DB2COMM شԫ "APPC"ٶDB2 Connect EE ʸ
      īܡ 24  39 
 
  6.  ٯ CS/NT 
 
      Ⱥ LU Ρȴ CS/NT SNA
      ɢ " LU6.2" عȺ LU  "" 
      ҳ "Yes"
 
  7.  ջܡȢ鬡 DIAGLEVEL ҳ 4
 
           update dbm cfg using diaglevel 4
 
      Ⱥƫ SPM عɡ

 
  8.  ٯ DB2
 
  9.   "db2diag.log"( DB2PATH  DB2INSTANCE
      ҳԶɱԫ롢ƫ SQLLIB\DB2 ̯Ⱥ)
      Ϊ SPM ɳٯƫ̯ɳٯعɡ
 
      ǩ SPM ġȹٯ롢ǡ "\sqllib\spmlog"
      硤ΪȺ㡢ع SPM ٯ
      Ь DB2  CS/NT 鬡
 
  DB2 Connect Enterprise Edition ʸīݷ仡
  Ⱥع
 
 
  13.3.1.1   SNA SPM ع
 
  1.   DB2 Щٯ CS/NTѡDB2 ȩٯ
      ʩЩѡ CS/NT ٯȺΪȴ DB2 Щٯ
      CS/NTDB2 ٯȩ衤DB2 ĸѡ CS/NT
      ٯش CS/NT ɢҳ Windows NT Τ٭Ȣٶ
      13.2.7, "CS/NT  Windows NT ĸ̥"
      ѥCS/NT ٯٯ
 
  2.   "DB2 Connect Enterprise Edition ʸī"  22 
      롢Microsoft ܬٯΤ¡ҡܬٯݨܡ (TP)
      ȢסȺMicrosoft ܬٯΤ¡롢
      ܬٯܡȢջ

 
  DB2 Connect Enterprise Edition ʸīݷ仡
  īȺع
 
 
| 13.3.2   AIX  4.1.4 ů 4.1.5 ĸ NIS ϯ
 
| ެ⡡  AIX  4.1.4 ̱ 5.2 
|        AIX  4.1.5 ̱ǿЩȴش祡
 
|  AIX  4.1.5  4.1.5ϯع˷ (NIS)
| š롢Ь̯ݨܡġ롢ȭ̯ع
| DB21015E
 
  ȺơݨܡѥȢ NIS Τ
   DB2 ϯ IDȺ

 
  1.  ڷ NIS Τƪϯ ID
 
  2.  ġĸ "/etc/passwd" 㡢ٴǡϯ
      ID㡢Ⱥϯ ID  NIS magic cookie
      Щmagic cookie 
 
           :!::::::
 
 
  13.3.3  DB2CPIC.DLL - WINDOWS NT ů WINDOWS 95 ĸ SNA 
          Ȣ
 
  ȴ SNA ˷ϯ SNA Ȣƫ
  ̯ĶġԶԶܨ¡
 
  o   DB2 Universal Database (UDB) for Windows NT  5
      (ȴԺ)
 
  o   DB2 Connect Enterprise Edition for Windows NT  5
 
  o   DB2 Connect Personal Edition  5 ( Windows 95 
      Windows NT ĸ)
 
  ġԶ DLLƵɢ db2cpic2.dll
   SNA Ȣ DLL  Fixpak 16for
  DB2 Universal Database  5 㡤 DLL db2cpic.dll
  ѡƷס̽ȹع(cmesi()) Ƶ
 
  +---------------------------------------------------------------------------+
  |  4. ġ db2cpic.dll                                        |
  +------------------------+-------------------------+------------------------+
  | Ϊ̦ȴȺ SNA ˷  |  WINDOWS NT       |  WINDOWS 95      |
  +------------------------+-------------------------+------------------------+
  | IBM Communications     | db2cpic2.dll ()       | db2cpic.dll ()       |
  | Server for Windows NT  |                         |                        |
  +------------------------+-------------------------+------------------------+
  | IBM Personal           | db2cpic2.dll            | db2cpic.dll            |
  | Communications for     |                         |                        |
  | Windows (ٶĶ    |                         |                        |
  | ެ)                  |                         |                        |
  +------------------------+-------------------------+------------------------+
  | Microsoft SNA Server   | db2cpic.dll             | db2cpic.dll            |
  +------------------------+-------------------------+------------------------+
  | Wall Data RUMBA (ȴ  | db2cpic2.dll            | db2cpic2.dll           |
  |  DB2 Connect     |                         |                        |
  | Personal Edition     |                         |                        |
  |  SNA )         |                         |                        |
  |                        |                         |                        |
  +------------------------+-------------------------+------------------------+
 
  Ϊ DB2 ī db2cpic2.dllƫĶ
  
 
       copy db2cpic2.dll db2cpic2.bak
     copy db2cpic.dll db2cpic2.dll
 
   db2cpic2.dll  db2cpic.dll ѡ
 
  ެ⡡Ϊ Windows 95 ĸů DB2 UDB
         5.0 IBM Personal Communications (PComm)
        for Windows 95  4.2 ( 4.1) db2cpic.dll
        ҳ db2cpic.bak db2cpic2.dll ҳ db2cpic.dll
         db2cpic.dll  db2cpic2.dll ѡ
 
 
  13.3.4   Windows NT  DB2NTMEMSIZE 
 
  DB2 UDB Extended Enterprise Edition for Windows NT ʸī
  㡢"̱ DB2 " ġȴݡ
  "DB2NTMEMSIZE  DB2 شٴĸ"
 
  Windows NT  DLL ȴشٴ
  ݨܡ١ DB2NTMEMSIZE
  ϯ Windows NT ĸ DB2 ԫǹӡ
  ԫܡ
 
  Ķشٴů̧
 
 
      شٴ                     
      ----------------------------------------------------------------------------
      ջȢ        16777216 (16MB)  DBMS:[]
      MPP FCM ٴ        33554432 (32MB)  FCM:[]
      ջܡ GUI        33554432 (32MB)  DBAT:[]
       DARI             16777216 (16MB)  APLD:[]
 
  ƫ ':' šġԶĸٴ

 
  ̱ DBMS ҳ () 256Ků̱ FCM ҳ () 64MB
  
 
      db2set DB2NTMEMSIZE=DBMS:256000:FCM:64000000
 
 
  13.4  ̱ SNAPSHOT Ȣ
 
   OS/2 snapshot Ȣʰҳ
  롢̱ƫܢǶعɡ

 
  ջģ́Ƕġǵ snapshot Ȣ롢
  ٯɢȡաɴȺջǵ (ջ)

  (ȺΪš޴ջ˷ӡЬߦš޴ջǵ
  ) Ϊƪջ A Ь
  ջ A紡ΪƷЬƫջ A

 
   (̧ǵ) ȴơڷܢ
  ̽ "ȴ"ơ
  Ȣ

 
 
  13.5  ڷ WEB ̽ DB2 
 
   Java  DB2 ( JDBC) عٶĸ
  Ķߡ

 
       http://www.software.ibm.com/data/db2/java/
 
 
  13.6   RS/6000 SP ĸٶ
 
  Ķع "IBM DB2 Universal Database Extended
  Enterprise Edition for AIX ʸī" ġ 6  " DB2
  Universal Database Extended Enterprise Edition"  1
  "ެ" ʢ㡤Ⱥٶ
  ȴ SP ĸ
 
   "ƫ no -a ٶܢ" ެ
  ѥĶ毡
 
  ĸԫٶ޲ԫΪٶҳԫЬ
  ٶҳĸԫ

 
   "Ӽ css0 " ҳƻѥĶ毡
 
  Ϊ /tftpboot/tuning/cst ˷䡢Ь
  Ķٶ

 
 
  13.7  š޴ջ CCA ̱
  (DB2 UNIVERSAL DATABASE EXTENDED ENTERPRISE EDITION FOR AIX)
 
  Ķع "IBM DB2 Universal Database Extended
  Enterprise Edition for AIX ʸī"  6  " DB2
  Universal Database Extended Enterprise Edition"  8
  "ڷ׻ɷȢ̱"
 
  ء޲ܡΤ¡ĸ discover=SEARCH
  ȺܡΤ¡ĸ̱ĸ

 
  ȴ̧Ȣƫ̶Ȣšڷ׻塤ƫ
  ڷ׻ɷȢ IMPORT/EXPORT ٯ̪šɢ
  ݣƶġԶ LoadLeveler ܨšڷ׻
  塤

 
  ΪĸȴܡΤ¡סЬƫ̱
 Τ䡢š޴ջ
  (ջš޴ٴΤ)

 
   TCP/IP ع̶ġ
 
  š޴ջ IPX/SPX Τߡ

 
 
  13.8  ڷ (UNIX)
 
   "ջƫ" 㡢 DB2 ѥ
  ެɳ "db2ckmig" ̦
  ջܡȢ㡢Ķ
 
       DB2DIR/instance/db2ckmig -e -a 0 -1 INSTHOME/migration.log
 
  ȩҳ
 
       DB2DIR/bin/db2ckmig -e -a 0 -l INSTHOME/migration.log
 
   " DB2 " 㡢 "db2imigr" 
  "-u fencedID" ɢҳƫٶ DB2 Client Application
  Enabler ѥ -u 紡̧ȴ DB2 ܨơ
  -u 
 
 
  13.9  DB2 SNMP ȹܡȢ
 
  ܡع̶ (SNMP) ܨ IBM NetView for AIX 
  IBM NetView for OS/2DB2  DB2 SNMP ȹܡȢ
  SNMP ܡܨΡԶȹܡȢīҳ DB2 Server ǵ
  ġSNMP ܡܨ NetViewթܡ˷ǵ
  DB2 SNMP ȹܡȢթܡ̧ SNMP ̿ܡġܡ
  DB2 Τ¡

 
   SNMP ȹܡȢĶ
 
  o    OS/2 ĸ
      -   TCP/IP  2.0 ĸ DPI 1.1 Ρ
      -    OS/2 Warp ġ Warp Connect  DPI 2.0 Ρ
      -   IBM SystemView Agent  DPI 2.0 Ρ
  o    Windows NT  AIX ĸ
      -   IBM SystemView Agent  DPI 2.0 Ρ
 
 
  13.10   CCA ڷ׻̯ DB2 Τ
 
  ҳħ IBM Communications Server SNA API Client for Windows
  NT  Windows 95 ١ٯǶ APPC 
  GUI ȢЩǿٯ APING Ȣ GUI Ȣ
  ġڷ׻ɷȢס̱סůǶ ODBC GUI
  Ȣ
 
 
  13.11  ܡ OS/2 ϯܡ
 
   "ϯܡ" ֪ 7 ġԶެȴȡ
  Ⱥެϯ ID ԫ 7. "ϯ ID
  " ެĶ
 
       ΪեܢȺ㡢Ь USERID  PASSWORD
       Ϊٶ README 
 
  Ⱥެȴ桢ܩơ
 
 
  13.12   SOLARIS ĸ EEE  ADSM ڷ׻ -  DB2PROFILE
 
   ADSM Client for DB2 UDB Extended Enterprise Edition for
  Solaris 롢DSMI_DIRDSMI_CONFIG DSMI_LOG ADSM 
   db2profile 㡤ΪƷ .profile  .kshrc
  Ьš޴ջġԶ Solaris ĸ
   ADSM ڷ׻ҳ 2.1.0.8 ꡤ
 
 
  13.13  ڷ DB2 FOR OS/390  5 ̯ DB2 UNIVERSAL DATABASE  5
 
  Ķع DB2 ɢ׻ɢҳȢ
  Τ硢 DRDA ȢΤ PTF 硨
 
       Ϊڷ DB2 for OS/390  5 ̯ DB2 Universal
       Database  5ſ DB2 for OS/390 APAR PQ07537
       ԺȢ

 
 
  13.14   WINDOWS NT Τĸ NETBIOS
 
   Windows NT ĸ NetBIOS ȭȭٴ衤
  紡ƫ DB2 شԫ db2nbrecvbuffsize ɬΤȺ
 
  Τĸȭٴҳڷ׻޲ء
   RQRIOBLK  32767 ڡΤ "ȭٴ
  "  4096 ڡٯ DB2 ЩĶ
 
       db2set db2nbrecvbuffsize=32768 -g
 
  ȺΤȭٴѡ
 
 
  13.15  ش - WINDOWS NT  WINDOWS 95
 
   Windows NT  Windows 95 ơʸīܡ "ܡ
  " 㡢ſĶԺع
 
  ջܡȢشƭWindows NT  x'20010000'
  Windows 95  x'90000000'ΪشЬƫԺ
  ԶԫڰջܡȢش
  ĸ DB2DBMSADDR شԫ̧Ķ
  Windows NT  x'20010000' ̯ x'B0000000'Windows 95 
  x'80000000' ̯ x'B00000000' x'4000000' ɢҳء
  Windows 95 ĸԫx'840000000'x'88000000'x'8C000000'...
 
  DB2DBMSADDR ԫĲ֪ȢЩȴ 0xƫ db2set
  ش
 
  Ӽ DB2DBMSADDR ҳ x'84000000'Ķ
 
       db2set DB2DBMSADDR=84000000
 
 
  13.16   DB2 JDBC APPLET Τɢ WINDOWS NT Τ٭Ȣ
 
   " Java Ȣ" - "Java Applets" Ķʸīܡ
  ϡܡīܡ̥Ķ
 
        appletsĶٯɢ
 
       1.  Web ΤĸīĶٯ DB2 JDBC applet Τ¡
 
         db2jstrt portno
 
         ̧ portno  DB2Applt.java  TCP/IP 

 
  Ⱥع롢ٶĶ
 
  DB2 JDBC Applet Server ܢƫٯҳ Windows NT Τ٭Ȣ
  ƫڷ̱ƺΤ٭ȢݷЩǿشҳΤ٭Ȣ

 
  شȺΤ٭ȢīҳܡϯĶ

 
       db2regjdbc  i
 
  شΤ٭Ȣϯ ID 硢硨
 
       db2regjdbc i [userid] [password]
 
  ƪΤ٭Ȣ硨
 
       db2regjdbc u
 
  ġشħΤ٭Ȣƫڷ̱ƺΤ٭Ȣݷٯϡ

 
  1.  ̱ƺ
  2.  ѺġĶΤ٭Ȣ
  3.   "DB2 JDBC Applet Server"
  4.  ī ( DB2Applt.java 
      ݷ TCP/IP )ѺġĶٯ
 
  Τ٭Ȣľƫҳٯٯϡ
 
 
  13.17  ̽ڵع
 
 
  13.17.1  ĸ
 
   "SQL "  SQL ȢΡȺ UNIX ɢ
  ˷ĸ
 
 
  13.17.2  DB2 
 
   DB2 ع֪͵ꦡġԶެɱ
  Ķ硨

 
                            ɱ
       ʨ                   A
                            J
                            K
                        C
                        T
 
 
  13.17.3   PostScript 
 
  Ϊ٣ƫ PostScript ͡
 
  ެ⡡ꢡ
 
   OS/2  Windows ƺĸ
 
  1.   PostScript ̯˷ĸ٫
       .exeԯ x:\doc\language\books\ps 硢̧
      x:  CD-ROM Ǽࡢlanguage 
       (EN  English)
 
  2.  ȺġܨΪ
      ġԶƫ PostScript ̧٫ҳ .psz
 
  3.  ȴİȹ 1 (ȹ) 
      PostScript 񢡤

 
  4.  ڷīĶ
 
           print filename.psz
 
   UNIX-based ƺ
 
  1.   CD-ROMٶʸīܡʩ CD-ROM
 
  2.  ̯ CD-ROM ĸ /cdrom/doc/%L/ps 硢̧ /cdrom 
      CD-ROM %L ꢡ
      ǿЩ̯㡢 .ps.Z ɢҳʢء
 
  3.  Ķ
 
      o    AIX ĸ
 
               zcat filename | qprt -P PSPrinter_queue
 
      o    HP-UXSolaris SCO UnixWare 7 ĸ
 
               zcat filename | lp -d PSPrinter_queue
 
      o    Silicon Graphics IRIX  SINIX ĸ
 
              zcat < filename | lp -d PSPrinter_queue
 
 
      ̧ filename  PostScript ٫ء
      PSprinter_queue  PostScript ꢡ
 
       AIX ĸ UNIX ʸīܡƫĶ

 
           zcat /cdrom/doc/en/ps/db2ixe50.ps.Z | qprt -P ps1
 
 
  13.17.4  ĸ
 
  ܨ (HTML) ֪Ȣ֪Ȣƫ
  ععḡľƫ
  ׻ջڵթ

 
  ƫǶ HTML  3.2 ݽ֪ĸ
 
  ĸ
 
  o   Ϊ DB2 ܡ̦Ьعס
 
  o    Web 䡤 DB2 
      ḡ

 
      -    UNIX-based ƺĸĶ
 
               file:/INSTHOME/sqllib/doc/%L/html/index.htm
 
          ̧ %L ꢡ
                __
 
      -   ̧ƺĸĶ
 
               sqllib\doc\html\index.htm
 
          Ⱥ DB2 ĸ
 
          ľƫѺ̥ĶDB2 ĸȺ
          ˷ӡȺܨ Windows 
          ĸ
 
          ެ⡡Ʒȴҡع롢DB2
                ĸ

 
 
  13.17.5  ǵΤ
 
  áDB2 ع˷ĸ̽ DB2 ع
  ġԶĩ DB2 عġ
  Ķ 
 
  1.  ڷ˷ĸ \sqllib\doc\html ȴȹ̯ Web
      Τ¡ġȴȹ硢ȩȺȴ
       HTML  GIF ѽȡ

 
  2.   web Τ¡عٶ
      Web Τĸ DB2 ĸǵס̧Ķ
 
           http://www.software.ibm.com/data/pubs/papers/db2html.html
 
  3.  Ϊ Java عƫҳȴ HTML  URL
      Ⱥ URL ̽ڵ̡

 
  4.  ġɻƫڦ

 
      o   
      o   ڦ
      o   ڦ ALTER TABLE 
      o   ֪
 
  عٶաġ
 
 
  13.17.6  ĸ
 
   HTML عƫĶٯɢ
 
  o    HTML Ƕġ͵ꦡѺġĶ DB2 ס
      ֪
 
  o    HTML Ƕġ͵ѺġĶſסſ
      
 
  o    HTML ſѥ Web 
      
 
  o    Web 䡢ʸ̯
 
  o   ع

 
 
  13.18  MICROSOFT INTERNET EXPLORER 4.0 HTTP 1.1 
 
  ЩInternet Explorer  DB2 Τ
  Ȣڷ Internet Explorer  HTTP 1.1 
  ԫť (ƫ̽ǵԶڷ׻ĸȺٯɢ) 
  ҳ Internet Explorer ̱ݨܡ˷
  񢡢桤
 
  1.  ̽Internet ס
 
  2.  ̯̽ȴťٴ
 
  3.  HTTP 1.1 ԫٴ proxy  HTTP
      1.1ů HTTP 1.1ť
 
  4.  ̽ס
 
  5.   Windows 񢡢ԫ桤
 
 
  13.19  ̱ڷܡ
 
  Զĸ DB2 UDB EEE for Windows NT ѥݨܡƷ
   "ȴ" ȩҳ̶Ȣڷڷ׻ĸꡤ
  ̱ڷܡġԶĸꡤ
  ȩȺĶ
 
  1.  ġĸ TCP/IP աġĸ
      ƫ 4000

 
  2.  ȩȺȴ
      ĸī
 
           DB2 update dbm cfg using svcename 4000
 
  3.  ġԶ̿ TCP/IPȴ
      ĸī
 
           rah "db2set db2comm=TCPIP"
 
  ġԶٯѥ̿ī TCP/IP ١
 
 
  14.0  REPLICATION вٶ
 
 
  14.1  ƺ
 
  IBM Replication ġٯŧ̦ڷ
  ʽ̯ĸƫĶ
 
  o    DB2 for MVSDB2 for VSE & VMDB2 Universal Database
       DB2 for OS/400 ջΤ
 
  o    MVSVMVSEOS/400AIXHP-UXSCO UnixWare 7
      OS/2Windows NTWindows 95Windows 98 Solaris 
      ɢĶƺ

 
  o    DB2 ĸ OracleSybaseInformix
       Microsoft SQLServer ( DB2 DataJoiner)
 
 
  14.2  ع
 
   21 IBM ʽعɡ㡢ħſȢ̥Զعɡ

 
       ASN1054S
       ſȢ̯ȴϯ "<src_ownr>"֪ "<src_tbl>"
       ܩ "<src_view_qual>" شع
 
       
       ֪ش졤
 
       ϯ
       ۹ششعش֪ɴ̱֪㡤
 
 
       ASN1055S
       ſȢ̯ȴϯ "<src_ownr>"֪ "<src_tbl>"
       ܩ "<src_view_qual>"ȴϯ "<tgt_ownr>"
       ֪ "<tgt_tbl>" ɴ̱ع
 
       
       ֪ش
 
       ϯ
       ۹ʽ
 
 
  14.3  DB2 UDB  5.2 ̽Ȣ LONG VARCHAR
 
  DB2 UDB  5.2 ̽Ȣܢ̦ȴ LONG VARCHAR 
  İ Long Field Manager شͷش
  ̽ LONG VARCHAR ع

 
  ̽ DB2 UDB  5 ϯơȴġԶ
  ƻ̯ ALTER TABLE ȢCREATE TABLE ȡ
 
  ɳ LONG VARCHAR ݨ ALTER TABLE Ȣĸ
   DATA CAPTURE CHANGES INCLUDE LONGVAR COLUMNS ƻ
  Ⱥ DDL Ȣ롢SYSCAT.TABLES  DATA CAPTURE CHANGES
  ҳ "L"ȴ INCLUDE LONGVAR ֪롢
  ̽Ȣīů̽ Long Field Manager ش硤ش
  ֪թѥ֪թ

 
  Ϊ LONG VARCHAR Ρ ALTER TABLE Ȣĸ
   DATA CAPTURE CHANGES ԫƷ "YES"
   "NONE" ڧ̧ƻ
 
  ش LONG VARCHAR ڷ̱Ķٯɢ
 
  1.   "ҳʽ"  "ƫ" 

 
  2.  ̽ LONG VARCHAR 
 
  ̱ DELETE ɢ (IBMSNAP_OPERATION="D")LONG VARCHAR
        ԫҳ NULL
 
 
| 14.4  DATAJOINER REPLICATION ADMINISTRATION TOOL README
 
|  DB2 DataJoiner Replication Administration Tool,  2.1.1
| (DJRA)
 
| ȴ DJRA ع
 
| o    DJRA
 
| o    DJRA
 
|  DJRA
 
| DJRA ġȹɢҳ DB2 UDB FixPak 7 for Windows 95Windows
| 98 ů Windows NT Ȣ롤Ʒɢ˷ӡ
| DJRA ʽܡ̪ DataPropagator Relational
| for MVS  5 ů DataPropagatorRelational  5 for AS/400 
| ܡ̪ơ DJRḀԶƺܡ̪㡢
| ȴġЩUDB ̱̽ڵ
 
|  DJRA
 
| DJRA  "\BONUS\REPLICATION" 㡤ع
| ƫĶ "DataJoiner Planning, installation and
| Configuration Guide" ̽ڵ
 
|      http://www.software.ibm.com/data/datajoiner/booksv2/djxn2m79
 
| ľƫԶɱ (ҳ SC26-915-01) postscript 
| 롢ڷ DataJoiner ftp ׻Ķ "v2ntpic.pdf" 
 
|      ftp://ftp.software.ibm.com/ps/products/datajoiner/info/
 
| Replication Administration  Windows 95Windows 98 
| Windows NT ɢ˷ĸ Replication Administration 
| 2.1.1 롢
 
| 1.   'ʼ'  'Explorer'
 
| 2.  ѺġĶ UDB DB2 CD-ROM
 
| 3.  ̽ BONUS
 
| 4.  ̽ REPLICATION
 
| 5.   DJRA Ȣʩ
 
| 桢ƫ DJRA 硤
ľƫ README.TXT
| ȴȢ޲ع
 
 
  15.0  DB2 Ȣв
 
 
  15.1   APPLICATION DEVELOPER'S KIT
 
  VisualAge for Basic  Lotus Approach ܨܵ
 
 
  16.0  SQL REFERENCE (ŧ)
 
 
| 16.1  NAMING CONVENTIONS (CURSOR-NAME)
 
| CURSOR-NAME         A long identifier that designates an SQL cursor. For host
|                     compatibility, it supports hyphens as well.
 
 
| 16.2  ALTER TABLE
 
  The ACTIVATE NOT LOGGED INITIALLY clause should not be used when DATALINK
  columns with the FILE LINK CONTROL attribute are being added to the table.
 
| Replace the description of "table-name" with the following:
 
| "table-name"
 
| Identifies the table to be changed. It must be a table described in the
| catalog and must not be a view or catalog table. If it is a summary table,
| alterations are limited to activating not logged initially, changing pctfree,
| locksize or append.
 
 
| 16.3  COMPOUND SQL
 
| The following should be added to the "Rules" section:
 
| DB2-Connect does not support SELECT statements selecting LOB columns in a
| compound SQL block.
 
 
  16.4  CREATE FUNCTION
 
 
  16.4.1  DBINFO
 
  Under both CREATE FUNCTION (External Scalar) and CREATE FUNCTION (External
  Table), there is a list of the information contained in the DBINFO structure
  which is passed to the UDF if the definer specifies DBINFO. The following
  bullet has been added to each of these lists:
 
       Application ID - unique application ID which is established
          for each connection to the database.
 
 
| 16.4.2  ERROR PROCESSING FOR UDF CALL TYPES
 
| In the "CREATE FUNCTION (External Scalar)" section, where the "NO FINAL CALL
| or FINAL CALL" specification is discussed, this sentence should be added to
| the bottom:
 
| A description of the scalar UDF processing of these calls when errors occur
| is included in the Embedded SQL Programming Guide, Chapter 7, "Other Coding
| Considerations" section, under "Hints and Tips".
 
| This updated description is in the section 8.11.2, "External UDF Error
| Processing Model for Scalar Functions".
 
| In the "CREATE FUNCTION (External Table)" section, where the "NO FINAL CALL
| or FINAL CALL" specification is discussed, this sentence should be added to
| the bottom:
 
| A description of the table UDF processing of these calls when errors occur is
| included in the Embedded SQL Programming Guide, Chapter 7, "Table Function
| Considerations" section.
 
| This updated description is in the section 8.11.1, "Scratchpad and Final Call
| Type Arguments for EXTERNAL TABLE Functions".
 
 
| 16.5  CREATE TABLE
 
 
| 16.5.1  CREATE TABLE DESCRIPTION
 
| Add the following to the description of NOT LOGGED INITIALLY:
 
  The NOT LOGGED INITIALLY clause should not be used when DATALINK columns with
  the FILE LINK CONTROL attribute are present in the table.
 
| Replace the description of "FULLSELECT" under "summary-table-definition" with
| the following:
 
| FULLSELECT
 
| Defines the query in which the table is based. The summary-table-options
| specified define attributes of the summary table. The option chosen also
| defines the contents of the fullselect as follows:
 
| When DEFINITION ONLY is specified, any valid fullselect can be specified.
| The data type and nullability attribute are determined from the result
| columns of the select list. Every select list element must have a name.
 
| When REFRESH DEFERRED or REFRESH IMMEDIATE is specified, the fullselect
| cannot include:
 
| o   references to a view, summary table, or typed table in any FROM clause
 
| o   expressions that are a reference type or DATALINK type (or distinct type
|     based on these types)
 
| o   functions that have external action
 
| o   functions that depend on physical characteristics (for example
|     NODENUMBER, PARTITION)
 
| o   table or view references to system objects (explain tables also should
|     not be specified).
 
| Furthermore, when REFRESH IMMEDIATE is specified:
 
| o   the fullselect must be a subselect
 
| o   the subselect cannot include:
 
|     -   functions that are not deterministic
 
|     -   scalar fullselects
 
|     -   predicates with fullselects
 
|     -   special registers
 
| o   a GROUP BY clause must be included in the subselect
 
| o   the select list must have a COUNT(*) function (or COUNT_BIG(*)) and no
|     DISTINCT
 
| o   only SUM (of not nullable columns), COUNT, or COUNT_BIG column functions
|     are allowed in the select list (without DISTINCT) and other select list
|     items must be included in the GROUP BY clause
 
| o   the FROM clause must reference a single base table (tables cannot be
|     summary tables, typed tables, or views)
 
| o   all GROUP BY items must be included in the select list
 
| o   no grouping sets are allowed (including CUBE and ROLLUP) or grouping on
|     constants
 
| o   a HAVING clause is not allowed
 
| o   if in a multiple partition nodegroup, then a partitioning key must be a
|     subset of the group by items.
 
| Replace the description of "REPLICATED" with the following:
 
| REPLICATED
 
| Specifies that the data stored in the table is physically replicated on each
| database partition of the nodegroup of the table space in which the table is
| defined. This means that a copy of all the data in the table exists on each
| of these database partitions. This option can only be specified for a summary
| table defined with REFRESH DEFERRED (SQLSTATE 42997).
 
 
| 16.5.2  CREATE TABLE NOTES
 
| Replace the seventh major bullet under "Notes" with the following:
 
| o   A REFRESH DEFERRED summary table may be used to optimize the processing
|     of queries. In order for this optimization to be able to use a summary
|     table, the fullselect must conform to certain rules in addition to those
|     already described. The fullselect must:
 
|     -   be a subselect with a GROUP BY
 
|     -   not include DISTINCT anywhere in the select list
 
|     -   not include any grouping sets (including CUBE and ROLLUP)
 
|     -   not allow nulls in any grouping columns in the group by clause
 
|     -   not reference any column more than once
 
|     -   not contain more than 32 columns in the group by clause
 
|     -   not include any special registers
 
|     -   not include functions that are not deterministic.
 
|     If the query specified when creating a REFRESH DEFERRED summary table
|     does not conform to these rules, a warning is returned (SQLSTATE 01633).
 
| Add the following new bullet under "Notes":
 
| o   If a summary table is defined with REFRESH IMMEDIATE, it is possible for
|     an error to occur when attempting to apply the change resulting from an
|     insert or update of the underlying table. The error will cause the
|     failure of the insert or update of the underlying table.
 
 
| 16.6  CREATE TABLESPACE
 
| The functionality described in the sub-section "DROPPED TABLE RECOVERY" has
| not been implemented. This sub-section should be deleted from Version 5.2 of
| the SQL Reference.
 
 
  16.7  APPENDIX A. SQL LIMITS/DATABASE MANAGER LIMITS
 
  The limit for the number of tables in a DMS table space has been increased
  from 6648 to 51971. This new limit will automatically apply to any table
  spaces with a page size other than 4KB, and to all newly created 4KB DMS
  table spaces.
 
  For existing 4KB DMS table spaces, the new /ets option of the db2dart utility
  is used to apply the new limit:
 
  1.  Deactivate the database (that is, ensure that no connections to it
      exist).
 
  2.  Invoke: db2dart <dbalias> /ets /tsi <tid>. This will check the specified
      table space for existing table, index, or LONG objects which have
      identifiers that are incompatible with the application of the new limit.
      If none are found, the new limit will be applied to that table space.
      Otherwise, a list of incompatible tables will be provided in the dart
      report.  In this case, those tables must be dropped, and the command run
      successfully, for the new limit to become effective. The tables can be
      recreated and reloaded after the utility has been run successfully on all
      logical nodes. On EEE installations, run this command first on one
      logical node that is a member of the nodegroup in which the table space
      resides. If incompatible tables are identified, drop them, and reissue
      the command. Once you have run the command successfully on one logical
      node, you must run it again on every logical node in the nodegroup in
      which the table space resides (either the db2_all or the db2dart_all
      utility can be used). For information about db2dart options, invoke
      db2dart without any options (the utility can be found in sqllib\misc).
 
  NOTES:
 
  1.  If db2dart /ets is successfully run on a table space, restoring a table
      space backup image taken prior to running dbdart /ets will no longer be
      supported.
 
  2.  If db2dart /ets is successfully run on any table space in a database,
      restoring a database backup image taken prior to running db2dart /ets,
      and rolling forward through the point-in-time when db2dart /ets was run,
      will not be supported.
 
  3.  A database backup must be taken immediately after successfully running
      db2dart /ets (all table spaces on which the tool is successfully run will
      be placed in backup pending state).
 
 
  17.0  SYSTEM MONITOR GUIDE AND REFERENCR (ŧ)
 
 
  17.1  EVENT MONITORS ARE CREATED WITH LOCAL SCOPE (ŧ)
 
  The syntax diagram for the CREATE EVENT MONITOR statement shows that by
  default, event monitors are created with global scope. This is incorrect.
  Event monitors are created with local scope by default.
 
 
  18.0  
 
 
| 18.1   (/֪)
 
| ʢެƫܩ紡"REFRESH IMMEDIATE" ܢƫħ
 
 
| 18.2   JAVA Ȣů APPLETS 
 
  ܢӼ SQL for Java (SQLJ) Ȣ Java
  Development Kit (JDK)  1.1.4 ۡ
 
  ȴشġ Java Ȣաʩ UDF Ȣ
  šȩڷ׻ΤȢƫڷ׻ĸϯ
  ڷ̽Τĸʩ UDFЩ̯̿Ȣ
  Ķ
 
  DB2STP         ܢΤȢƵڷ׻Ȣ DB2SPCLI
 
  DB2UDF         ܢΤȢƵڷ׻Ȣ DB2UDCLI
 
  STP            ܢΤȢƵڷ׻Ȣ STPCLI
 
  UDF            ܢڷ׻ȢƵΤȢ UDFSRV
 
   UNIX ƺĸǡ Script  "embprep" Windows ƺĸ
  ǡ "embprep.bat" OS/2 ĸǡ "embprep.cmd"Щ
   SQLJ Ȣ "makefile" Ƶ Java  SQL 
  Ȣĸ "db2profc" թջϯ ID 
  ȴƫſȺɷڷ׻ڷ̯ DB2 Τĸ
  ջ
 
  Ķ DB2 UDB  5.2  SQLJ ̧̱
 
  o   Щ "DATETIME"
      ֪Ȣ

 
  o   Щ "PACKAGE USING package-name"Щ
      ǡꢡΪīꢡЬ (
      ٫ȩ)޲ 8 Զҳ SQLJ 
      ȴ _SJProfileN ʢ (Ⱥݨ N )
      ġ 8 Զۡ
      Щ (8 - pfKeyNumLen) Զȩ
 
      ̧ pfKeyNumLen Ϊ
       7Ь޲ѥ 7 ǿ
      ˡ
 
                                         
            ---------------------                  --------------------
            App_SJProfile1                         App_SJP1
            App_SJProfile123                       App_S123
            App_SJProfile1234567                   A1234567
            App_SJProfile12345678                  A2345678
 
  o    java.math.Bigdecimal Java Ȣ SQL Ȣ
      롢ȢЩ
      ΪڷȢĸĶ
      Ьƫ CAST 
 
  o    java.math.BigInteger  Java ɢҳ SQL Ȣ
      

 
  o   ȴ SQLJ ʩ FENCED ȢĶ硤
 
| o    sqlj  "profile" sqlj ܨ롢
|      "db2profc"
 
| o    "SET TRANSACTION" ƻ
 
| o    "WITH" ƻ
 
| o    SQLJ Űϯ JDBC ΪǨɢ̡
 
| o   Java Ȣ SQL Ȣ
 
  ߎڷ applet īǵ 
  īߎ applet 롢ȭ̯Ķعɡ
 
       java.lang.ClassNotFoundException: Applt_SJProfile0
 
  ƶȴġԶȢƫҳ Java ɱ֪Ȣ
   Ȣҳ
  sqlj.runtime.profile.util.SerProfileToClass  Java ɱ 
  ҳīȴ Java ɱ
  ܨҳ ƫĶ
 
       profconv Applt_SJProfile0.ser
 
         
 
       java sqlj.runtime.profile.util.SerProfileToClass Applt_SJProfile0.ser
 
  Ϊǡɱ Applt_SJProfile0.class .class ֪Ȣ
  ̽ applet  .ser ֪Ȣ
 
   SQLJ applet db2java.zip ů runtime.zip 
  Ϊȴ applet ɱdb2java.zip ů runtime.zip
  ġ Jar 㡢 db2java.zip ů runtime.zip (
  ) ī "applet" ٶ㡤 
  Զ zip ߎ¡ db2java.zip
  ̦ȴ applet ɱ runtime.zip  web ߎƫ̽
  ɢ㡤
 
 
  18.3  WEB ̱ NETQUESTION ( WEB ̱ǵ)
 
   UNIX ɢ˷ĸǵ
  ƫ Web Ķ׻̯
 
       http://www.software.ibm.com/data/pubs/papers/db2html.html
 
   OS/2  Windows ɢ˷ "nqmap" ҳ

 
       nqmap -u DB2ADMEN "http://yourserver/doc/html/" DB2ADMEN
 
   OS/2 ɢ˷ĸҡٯ HTML Τ¡Ķ

 
      1.  DB2 for OS/2 㡢ƭѺġĶٯ HTML
         Τ¡
         ̽סѺġĶȢ
         ƫٶ㡢 CONFIG DB2WEBCC.CNF ̯
         ٶʢѺġĶ OK ơ
      2.  db2init.cmd Ⱥ x:\sqllib\bin\̧ x: 
          DB2 񢡤Ⱥ "DB2SS.EXE" 
         "DB2NETQD.EXE" 硤Ⱥʢ CONFIG "DB2WEBCC.CNF"
         (ſ)֡
         
         ģ́ǹ

 
  Ϊ DB2 ĸع DB2 ǵ (
  UNIX ˷ĸҳ "ܨջ")  Web ̱ů̧ĸ멡
  Ϊ Web ̱ů̧멡̧ DB2 ĸǵ
  Ӽ̽ܡȢӡ DB2 Connect ˡЬ
  ܨ 24ſ⡢Ϊȴ
   DB2 ĸǵ Web ̱Ь Web ̱
  ĸܨ

 
 
  18.4  ̱̱
 
  Ķҡ̱̱
 
  o   ̱ش㡢 audit_buf_sz ٶ

 
  o   ջĶǵʩ (ϯڡ
      ֪Ϩ)ܢѺʩʩ
 
  o   š޴ջġԶĸ̿ ( DB2
      UDB EEE)
 
 
| 18.5  Web ̱ JAVA DEVELOPMENT KIT ȹ
 
| ̡ġ 45 ĸWeb ̱ȴݡ
| ܩĶƻů Java Development Kit ȹ
 
| "ڷǶ Java Development Kit 1.1.5  Jaba  Web
| 硤"
 
|  web ޲ů Java Development Kit
| ȹٶĶ web 
 
|      http://www.software.ibm.com/data/db2/udb/webcc/
 
 
  19.0  OS/2  HTML Τ¡       HTML ǵ
 
 
  IBM DB2 Universal Database (DB2 UDB) ĸǵ HTML 
  Τĸ̶ɷع OS/2 ĸ
  ˷ӡ

 
 
  19.1   OS/2 ĸ HTML ˷
 
 
  19.1.1 OS/2 ĸ˷ǿ۾ǵ
 
  ĸ TCP/IP  3 ꡤ   OS/2 Warp V4
  ҳ̦ެ TCP/IP ȹΪ OS/2 Warp V3 Internet Access Kit
  (IAK) ҳٴ (local loopback)ȴ޲ FixPakЬ OS/2 Warp V3
  Internet Access Kit (IAK) ǩȴ桤
 
   Netscape 2.02 for OS/2 ¡ Ϊ Netscape 
  ȩơЬ Web Explorer 1.1 ꡤ 㡢
    localhost  proxy ݨܡ
 
  Ϊ̦ȴ VisualAge for C++ for OS/2 ˷ĸ DB2 UDB
  Ь VisualAge for C++ ܨ¡ CSD6 ꡤ
 
 
  19.1.2   OS/2 ĸ TCP/IP
 
  Ʒ˷ĸȴ TCP/IP ٴ localhostЬȴ
  ȴƦΤƫɢ
 
  ٴӡ
 
  1.  OS/2 TCP/IP衤
 
  2.  TCP/IP 鬡شۡ
 
  3.  
 
  4.   ""  ""
 
  5.  Ϊ̽ "" ťܢ̽ơ
 
  6.    "IP "  127.0.0.1  "ȹ" Ϩ
 
  ˷ĸ localhost
 
  1.   localhost OS/2 ī
        "ping localhost"
 
      o   Ϊȴ衢ȴ localhostƫܩĶ
           2  3̯ 4
 
      o   Ϊ "localhost ϡ"Ь
          localhost 2
 
  2.  Ϊĸȴ (loopback) (ٶĸ
      "ٴ")
 
  3.  ΪĸĶ localhost
 
      a.   MPTN\BIN\SETUP.CMD 㡢Ķ̧̯
          ifconfig ѥ
 
               ifconfig lo 127.0.0.1
 
          ެ⡡Ӽ̦ެ Internet Access Kit  OS/2 Warpƫ
                Ⱥ̯ \STARTUP.CMD  MPTN\BIN\SETUP.CMD
                ΪȺ㡢Ьǡơ
 
      b.   TCP/IP 㡢Ķ
 
          1)   "εΤ٭Ȣ"
 
          2)   "ȴΤ" ֪㡢
              ġ硢 "IP " ҳ 127.0.0.1
               "" ҳ localhost
 
          ެ⡡Ϊ " LAN εΤ٭Ȣ" ĸ
                ȴꢡЬ IP  127.0.0.1
                ҳ localhost 롢Ⱥɢҳɱء
 
      c.  ̽ "ΤЩǿ HOSTS " 
 
          ެ⡡Ⱥ OS/2 ( localhost) 롢
                ĸ̯ߡΤ¡ 
                ΪĸOS/2 
                Τ檡
 
      d.   TCP/IP 鬡ѥٯ˷ӡ
 
      e.  ƫ ping localhost ̯ĸ
 
  4.   OS/2 ĸī "hostname"
       TCPIP شꢡ
      ĸꢡ 32 ԶΪ
      ۾ǵЬĸơ롢 CONFIG.SYS
      (SET HOSTNAME=<correct_name>) ơ
      ѥٯ
 
 
  19.1.3   TCP/IP 
 
  ѺġĶ "ٯ HTML Τ" Ⱥ DB2 for OS/2
  衤ΪܢعɡЬ TCP/IP
   19.1.2, " OS/2 ĸ TCP/IP"  TCP/IP
  ΪЬٯ OS/2

 
 
  19.1.4  ˷
 
  ˷Ķҳƫݷ̧ܨ
  Ϊ G  DB2 UDB ů˷ӡѥĴ
   H  VisualAge for JavaЬƷġԶ˷ -- ġԶ
  ˷ӡ
 
  ǵȺ Ⱥ硢
  OS/2 īȺ
 
       echo %IMNNLPSSRV%
 
 
  19.1.5  Τ
 
  Τ 49213Ⱥ TCP/IP 
  աΪȴƶġԶܨȺաƫɳĶ
  Τա

 
  1.  ˷ HTTPD.CNF ҳϡƫ
      ޲ 49000 ĸ
 
  2.   DB2 UDB 硢ѥ %DB2PATH%/doc/html 
       DB2SRCH.HTM Ķġ硢ȩҳ
      ǿЩ̽ա
 
           <form action="http://localhost:49213/cgi-bin/db2sr*xx.exe" method="POST">
 
           ̧㡨
              *        D  DBCSS  SBCS
              xx      ǵɱ
              49213    localhost 
 
  3.  ٯΤ¡ DB2 for OS/2 Ѻ
      ̥Ķ
 
 
  19.1.6 Τ
 
  ΪΤȭ̯عɡڷ OS/2 ī
  Ķߧ
 
       SNIFFLE /P
 
  ȩѥȴٯ˷ӡ
 
 
  19.2   HTML ˷
 
 
  19.2.1 ٯΤ
 
  ˷ѥѥٯٯ
 
  ΪΤٯȴ먡ܢĶ
  عɡЬѺġĶ " HTML Τ" ѥѺġĶ "ٯ
  HTML Τ" 
 
       먡̯Τ¡
       Τƫĸ
       ѥġȹ
 
  ȩѥΤȭشաơ
  Ѻ̥ĶDB2 for OS/2 " HTML Τ" 
  
 
 
  19.2.2 Ţ
 
  Ϊĸ (Ϊĸ檡ڷ LAN
  )Ь localhost ǵȺɢ
  ش 19.1.2, " OS/2 ĸ TCP/IP" 㡤
 
 
  19.2.3  ̯
 
  ˷ǿſ DB2 UDB ܨ¡
  ſ˷ҡ Ⱥ롢ȴ
  Ϊ "̯" عɡҳΪ
  ˷ĸܨ¡
 
 
  19.2.4  Ϊ ( 500 )
 
  ΪΤ¡ǵܢ먡ЬĶ
 
  1.  ˷ IMNNLPSSRV 
      ˷ (ٶ "19.1.4 ˷")
 
  2.  ˷
 
      o   DB2SRSxx.EXE  DB2SRDxx.EXE ƫ (Щϯ SBCS
          ǵѥϯ DBCS ǵ)xx ǵ
          ɱDB2SRSEN.EXE  SBCS ǵƫ
          DB2SRDCN   DBCS ǵƫ
 
      o   DB2HEAD.HTM ů DB2FOOT.HTMΪ㡢Ь
          DB2NETQ.CMD ſشů̯˷㡤

 
  3.  ˷ܨǵشƵȺ˷
      شǵ
 
           NQMAP -A
 
      DB2 UDB ǵҳ DB2ADMxxDB2APDxx DB2CONxx̧
       xx ǵɱܢ
       NQMAP 㡤
 
      Ϊ㡢Ьƫ TCP/IP ҡſ
      䴡Щϯƫ SNIFFLE /P ѥϯƫ
      ǵ
 
 
  19.2.5 ٯΤ٭Ȣ̯
 
  ΪܢȺ먡
 
       HS0410: ٯΤ٭Ȣ̯먡Τ٭Ȣ
               ٯơѺ Enter ...
 
       (ȴԷȺ EHS0411ѡ
       Ⱥعܢĸ)
 
  ڷѥƫաشơ
  NetQuestion 硢ī
 
       type netq.cfg
 
   NetQuestion ꢡΪꢡ
   DHCP  DDNS 롢Ьī
 
       netqinit <data-path>
 
       Example:  netqinit d:\netqos2\data
 
 
  19.3  HTML ˷
 
   DB2 for OS/2 ѥ˷ĸ
  UNINSTNQ.CMD ƪơ
 
 
  19.3.1  OS/2 ĸ˷
 
  Ϊ UNINSTNQ.CMD ˷ҡЬƫҡ

 
  o   TCP/IP 鬡
 
  o   DB2 UDB ƶġԶܨ˷ӡȺƪơ
 
  1.   147 ĸ 19.1.2, "  OS/2 ĸ TCP/IP" 
       TCP/IP TCP/IP 
      Τ¡Ϊȴ鬡ġٯ˷ӡ
 
  2.  ɲܨ˷شĶ
 
           NQMAP -A
 
      ΪȺǶſЬ̯ 6
 
      ΪȺ DB2 UDB ſ (ǩɻ̧
       "DB2")ЬƪΤ¡
 
      ΪȴǶ DB2 ſ (DB2ADMxxDB2APDxxDB2CONxx)
      Ь DB2 UDB ּ̽شſӳ˷Ϊ
      DB2 UDB ҡĸʰ (SQLLIB ݷ
      ɴ)Ķٯּ̽شſĶ
      ƪ˷硤
 
      a.  ȺΤ硨
 
               NETQ START SERVER           //for SBCS
               IMQSS.EXE -START DBCSHELP   //for DBCS
 
      b.  ġԶſȺ
 
               NQMAP -D <index_name>
 
      c.  ġԶſȺ
 
               NQDELET <index_name>        //for SBCS
               TMDELET <index_name>        //for DBCS
 
      d.  Τ¡
 
               NETQ STOP SERVER            //for SBCS
               IMQSS.EXE -STOP DBCSHELP    //for DBCS
 
      e.   "NQMAP -A" ȴĶǶ DB2 ſ (DB2xxxxx)
          Ϊȴ IBM Τ٭ĩޡ
 
      f.  Ķ̧ſɢ㡤
 
               NQCOUNTI <search system directory>  //for SBCS
               TMCOUNTI <search system directory>  //for DBCS
 
          ެ⡡Ϊϡٶ 19.1.4, "˷
                "
 
          o   ΪȺȴġԶԶſɢ㡢
              Ь˷ƪҳ̧ܨش㡤
              ̧
 
          o   Ϊ˷ "0 ſɢ"Ь UNINSTNQ.CMD
              (ܩȺعɡ"SYS0016: Ⱥݷƪ") Ⱥ
              ˷ݷƪΪ㡢Ь IBM Τ٭ĩޡ
 
      g.  UNINSTNQ.CMD ȩѥĶ
 
          1)  ڷǡެǹߦ CONFIG.SYS
 
              ެ⡡ǡެǹ config.xyz ֪Ȣ̧
                    xyz  000 ̯ 100 ġԶƫ
                     ެǹ̦
                    ȴشȴḁ̇̄Զެǹ 
                    ġԶ DB2 UDB ҡġ˷ҡ
 
              o   ӼƷƪ˷ĨԶެǹ
 
              o   Ӽƪ DB2 UDB ˷ġԶެǹ
 
          2)  ƪ˷ů̧ȴ硤
 
              a)   %TMP%/NETQ 硤
 
              b)  ɴ NETQ ޷
 
              c)  ɴ NETQ 硤
 
              d)  ٯ
 
 
  20.0  WINDOWS NTWINDOWS 95 ů WINDOWS 98  HTML Τ¡
         HTML ǵ
 
 
  IBM DB2 Universal Database (DB2 UDB) ĸǵ HTML 
  ˷ĸ̶ɷع  Windows NT
  Windows 95 Windows 98 ĸ˷ӡ
 
  ˷ſΤȩ˷
  Ķҳƫݷ̧ܨ
 
 
  20.1  ̱
 
  Ⱥ˷ Windows 3.1
 
  Ⱥ˷ Thin Clients  Ϊ̧ǵȭ̯
  먡
 
 
  20.2  WINDOWS NTWINDOWS 95 WINDOWS 98 ĸ HTML ˷
 
 
  20.2.1  ˷ǿ۾ǵ
 
  Ķǿ۾ǵ
 
  o    Windows NT 4.0ڵ̯޲Ϊ Service Pack 3
      عٶ
      http://www.microsoft.com/ntserver/info/servicepack3.htm
 
  o    Netscape 3.0 ¡㡢
       localhost  proxy ݨܡ
 
  o   ĸȴ TCP/IP  3 ꡤơ
       DB2 UDB ˷ɢ
 
       Windows 95ĶȢ TCP/IP
 
      -    LAN Ʀơ
 
          -   ȴ DNS
 
          -   LAN DNS  "localhost" εҳ 127.0.0.1
 
          -   Ţ LAN ƦĶ硤
 
      -   Ʀơ
 
          -    DNS
 
          -   ٯ̽ڵ TCP/IP ߡ
 
              ެ⡡ɻƷġԶ TCP/IP Ʀա
                    ȴ TCP/IP ƦӼ
                    ȴ鬡 LAN 
                    Ʀ
 
          Τ٭׻ (ISP)  TCP/IP ﳡ
           ISP 鬡 TCP/IP ﳡ
          Ʀ TCP/IP ﳡѥϯ Windows 95 ̱ƺ
           "" 鬡Ʀ TCP/IP 
          ĸȢ鬡СƦ TCP/IP
           DNSƦ TCP/IP  IP
          ߡ ISP 鬡Ⱥɭ
 
       Windows NT 4.0 ơĸ̥ TCP/IP ȴ桤Ϊ
      ĸ硢Ь̧̥ԶƦľƫ MS Loopback
      Ʀ
 
 
  20.2.2 ǿЩ˷
 
  ΪƶġԶܨǿЩ˷ ( VisualAge for Java)Ь
  Τ¡Τ¡ѺġĶ
  DB2  " HTML Τ"
 
 
  20.2.3  ˷
 
  ˷Ķҳƫݷ̧ܨ 
  Ϊ G  DB2 UDB ů˷ӡѥĴ
   H  VisualAge for JavaЬƷġԶ˷ -- ġԶ
  ˷ӡ
 
  ǵȺ Ⱥ硢
  Ⱥ
 
       echo %IMNINSTSRV%
 
 
  20.2.4 ҳΤ̧
 
  Τ 49213Ⱥ TCP/IP 
  աΪȴƶġԶܨȺաƫɳĶ
  Τա

 
  1.  ˷ HTTPD.CNF ҳϡƫ
      ޲ 49000 ĸ
 
  2.  Ķɲ <db2path> 
 
 
           DB2SET DB2PATH
 
  3.   DB2 UDB 硢ѥ <db2path>/doc/html 㡢
       DB2SRCH.HTM Ķġ硢ȩҳ
      ǿЩ̽ա
 
           <form action="http://localhost:49213/cgi-bin/db2sr*xx.exe" method="POST">
 
           ̧㡨
             *        D  DBCSS  SBCS
             xx      ǵɱ
             49213    localhost 
 
  4.  ٯΤ¡Τ롢ѺġĶ
       DB2  " HTML Τ"ٯ
      Τ롢ѺġĶ DB2 
      "ٯ HTML Τ"
 
 
  20.2.5   LAN ĸ DB2 UNIVERSAL DATABASE
 
  Ϊ Windows 95 ĸ LAN ĸ DB2 UDB
  Windows 95 ĸ AUTOEXEC.BAT ЩȺ㡢˷
  ̧˷ȹ (IMNENV.BAT)
  Ⱥȹ˷ (F:\IMNNQ_95)
  Զڷ˷ IMNENV.BAT ̯ƶġԶ
   AUTOEXEC.BAT Щ/硤
  ѥԺ
  AUTOEXEC.BAT ٯƵȺȹΪ IMNENV.BAT
  ̯ C:\WINDOWS\IMNNQ 㡢ƫĶġ̯ AUTOEXEC.BAT 㡨
 
       IF EXIST C:\WINDOWS\IMNNQ\IMNENV.BAT CALL IMNENV.BAT
 
 
  20.2.6 Τ
 
  Ϊ DB2 UDB ΤЬ
  DB2 UDB ɢȩĶ̶ɷ̽
  ٯɢ
 
   <temp>\imnnq\install  (̧ <temp> ˷ %TEMP%
  ),  IMNNQ.ERR Ϊ㡢Ь
  ܨ¡Ϊ IMNNQ.ERR 㡢ЬĶƫܢթ
 
 
  HTML Search Server for Windows NT, Windows 95, and Windows Documentation H157
 
 
       1 - Ⱥܢ PATH ŴۡΤ̯ PATH 㡢ӳ
           Զ PATH ݷּ桤
           ެ⡡  Windows NT 4.0 ̱ 512Windows 95 ̱
            255
           Ķ
           a)  AUTOEXEC.BAT  PATH  (PATHGOOD)
           ѥ񢡤
           b) ڷ <temp>\imnnq\install ƪ IMNNQ.ERR 
           c) ܨҡ˷ӡ
           d)  PATHGOOD  PATH ѥϯ޲ɢ
           ǡ
 
       2 - 먡 IBM Τ٭ĩޡ
 
       3 - Ϩ먡ŷȴ 4.5MB Ϩ˷
           ȴϨ Windows 95  AUTOEXEC.BAT ա
 
  Ϊ IMNNQ.ERR ġعɡ XXX.EXE 㡢Ь̯˷ƫ
  ˷/Ȣٶ 20.3.4, "
  Ϊ ( 500 )"
 
 
  20.3   HTML ˷
 
 
  20.3.1  ٯΤ
 
  ˷ѥѥٯٯ
 
  ΪΤٯȴ먡ܢĶ
  عɡЬѺġĶ " HTML Τ" ѥ
   DB2 ѺġĶ "ٯ HTML Τ"
 
       먡̯Τ¡
       Τƫĸ
       ѥġȹ
 
  ȩѥΤȭشա롢
  ѺġĶ DB2  " HTML Τ
  "
 
 
  20.3.2  NETSCAPE  INTERNET EXPLORER  PROXIES
 
  Ϊ Netscape  Internet Explorer ٯ proxiesЬƫ
  Ժ proxy ع
 
        Netscape 3
       --------------
         1. ̽ - 
         2. ѺġĶ Proxies 
         3. ѺġĶٯ Proxy 鬡ס
         4.  "No proxies for" ī
              localhost:49213
            ΪȺݨȴ̧硢ٴҡ
         5. ѺġĶסٯ Proxy 鬡᧡
         6. ѺġĶס֡᧡
 
        Netscape 4 (Communicator) 㡨
       -----------------------------
         1. ̽ - 
         2. ˡѺ̥Ķס
         3. ѺġĶProxiesס
         4. ѺġĶٯ Proxy 鬡ס
         5.  "... Ⱥҳꢡ
            Ŧ proxy Τ" ī
              localhost:49213
            ΪȺݨȴ̧硢ٴҡ
         6. ѺġĶסٯ Proxy 鬡᧡
         7. ѺġĶס֡᧡
 
        Internet Explorer 3 㡨
       -----------------------
         1. ̽ - ס
         2. ̽١ס
         3.  "... Ⱥҳꢡ
            Ŧ proxy Τ" ī
              localhost:49213
            ΪȺݨȴ̧硢ٴҡ
         4. ̽ "ٴ (intranet) Ŧ proxy Τ" 
         5. ѺġĶס֡᧡
 
        Internet Explorer 4 㡨
       -----------------------
         1. ̽... Internet ס
         2. ̽١
         3. ̽ٴ (Intranet) ̽ܩ proxy Τ¡ť
            ެ⡡ Ϊ proxy  socks ١̽
            Proxy Τ̽ InternetťȺť
         4. ̽ "" Ѻڡ
         5. ī
              localhost:49213
             "... ȺҳߡŦ proxy Τ
            " 
            ΪȺݨȴ̧硢šٴ硤
         6. ̽ס֡
 
 
  HTML Search Server for Windows NT, Windows 95, and Windows Documentation H159
 
 
  20.3.3 ĸ WINDOWS 95 
 
  Τ¡
 
  ΪĸΡġ̯ LAN ɢҳŢƺЬ
  ƫȴšɳ TCP/IP 
  Τ IP ߡƻ멡̥Զ TCP/IP --ġԶ
  ɢƶġԶŢɢ
 
   Windows 95  TCP/IP ޲ԺȢ
 
  Windows 95 Ʒթȴġ TCP/IP 鬡 紡ĸȴ
  Ȣƫġĸѥ
  ԯ ( LAN Ţ) TCPSwitch Ȣ̧ġ
 
  Socks Τ proxies
 
  Ϊ Netscape 3īߡ proxy  socks Τ
  ̽ internetƫЩ Netscape ɴ
   Netscape 3 ġԶ--Netscape Communicator  "
  " ƫԺȺ
 
 
  20.3.4  Ϊ ( 500 )
 
  Ϊܨȴ桢桢ЬĶ

 
  1.  ˷ IMNINST 
      IMNINSTSRV IMNINSTSRV ˷ (ٶ
      20.2.3, "˷")
 
  2.  ˷
 
      o   DB2SRSxx  DB2SRDxx ƫ (Щϯ SBCS ǵ
          ѥϯ DBCS ǵ)xx ǵ
          ɱDB2SRSEN  SBCS ǵƫ
          DB2SRDCN   DBCS ǵƫ
 
      o   DB2HEAD.HTM ů DB2FOOT.HTM
 
  3.  ˷ܨǵش
      ƵȺ
      ˷شǵ
 
           NQMAP -A              //for SBCS
           TMMAP -A              //for DBCS
 
      DB2 UDB ǵҳ DB2ADMxxDB2APDxx DB2CONxx̧
       xx ǵɱܢ
       NQMAP ( TMMAP) 㡤
 
       NQMAP ( TMMAP) ſƫĶ
      ̽ڵġԶſ̧ء
 
           IMNIXSTA <index_name>  //for SBCS
           IMQIXSTA <index_name>  //for SBCS
 
      Ⱥءſſǵ
 
      Ϊĸġ١ƫ DB2 UDB ܨȢ
      Ϊġŷ DB2SRCH.EXEDB2HEAD.HTM
      DB2FOOT.HTMЬƫڷ <db2path>\misc ̯˷
       (E:\IMNNQ_NT)ܨȢΤ
      
 
 
  20.3.5  ̯
 
  ˷ǿſ DB2 UDB ܨ¡
  ſ˷ҡ Ⱥ롢ȴ
  Ϊ "̯" عɡҳΪ
  ˷ĸܨ¡ǵ
 
 
  20.3.6  ٯΤ٭Ȣ̯
 
  ΪܢȺ먡
 
       EHS0410: ٯΤ٭Ȣ̯먡Τ٭Ȣ
               ٯơѺ Enter ...
 
       (ȴԷȺ EHS0411ѡ
       Ⱥعܢĸ)
 
  ڷѥƫաشơ
  NetQuestion 硢ī
 
       type netq.cfg
 
   NetQuestion ꢡΪꢡ
   DHCP  DDNS 롢Ьī
 
       netqinit <data-path>
 
       Example:  netqinit c:\imnnq_nt\data
 
 
  HTML Search Server for Windows NT, Windows 95, and Windows Documentation H161
 
 
  20.4   HTML ˷
 
  ެ⡡ DB2 UDB ܨЩ˷ϡ
          ˷ӡѺġĶ DB2 
          " HTML Τ"
 
  ˷ DB2 UDB ܨҡ̧ɢѺġĶ
   DB2  ""
 
 
  20.4.1 Ϊ˷ڷ WINDOWS NTWINDOWS 95  WINDOWS 98
         
 
  Ϊ˷ҡ̧ҳ
 
  o   DB2 UDB ƶġԶܨ˷ӡȺƪơ
 
  o   ˷硤
 
  ɲܨ˷شĶ
 
       NQMAP -A            //for SBCS
       TMMAP -A            //for DBCS
 
  ΪȺǶſЬ̯ 6
 
  ΪȺ DB2 UDB ſ (ǩɻ̧ "DB2")
  ЬƪΤ¡
 
  ΪȴǶ DB2 ſ (DB2ADMxxDB2APDxxDB2CONxx)
  Ь DB2 UDB ּ̽شſӳ˷Ϊ
  DB2 UDB ҡĸʰ (SQLLIB ݷ
  ɴ)Ķٯּ̽شſĶ
  ƪ˷硤
 
  1.  ȺΤ硨
 
           IMNSS START SERVER          //for SBCS
           IMQSS.EXE -START DBCSHELP   //for DBCS
 
  2.  ġԶſȺ
 
           NQMAP -D <index_name>       //for SBCS
           TMMAP -D <index_name>       //for DBCS
 
  3.  ġԶſȺ
 
           NQDELET <index_name>        //for SBCS
           TMDELET <index_name>        //for DBCS
 
  4.  Τ¡
 
           IMNSS STOP SERVER           //for SBCS
           IMQSS.EXE -STOP DBCSHELP    //for DBCS
 
  5.   "NQMAP -A" ( "TMMAP -A") ȴĶǶ DB2 ſ
      (DB2xxxxx)Ϊȴ IBM Τ٭ĩޡ
 
  6.  Ķ̧ſɢ㡤
 
           NQCOUNTI <search system directory>  //for SBCS
           TMCOUNTI <search system directory>  //for DBCS
 
  ެ⡡Ϊϡٶ 20.2.3 "˷"
 
  o   ΪȺȴġԶԶſɢ㡢Ь
      ˷ƪҳ̧ܨش㡤Ŧ̧
 
  o   Ϊ˷ "0 ſɢ" UNINSTNQ.EXEΪ
      ˷ҡƫٯȢƪܨ¡
 
      1.  ƪĶĶش硨
 
               \\HKEY_LOCAL_MACHINE\SOFTWARE\IBM\NetQuestion
 
          ƪȺ硢̧ȴ硤
 
      2.  ƪ˷ů̧ȴ (D:\IMNNQ_NT)
 
      3.  ƪ IMNINST  IMNINSTSRVڷ PATH 
          ƪ˷Ϊ˷ӡ
          IBM Τ٭ĩޡ
 
 
  20.4.2 ܨѥ
 
  DB2 UDB ѥƶġԶЩġ񢡤
  ȴ˷ DLL ݷɢ˷硢̯Ķġȹ
  ƪΪЩ˷ҡЬ˷
  DLL Ķġȹݷɴ桢ܨ˷
 
 
| 21.0  NETQUESTION 
 
 
| 21.1   UNIX ƺĸ IMNSEARCH (NETQUESTION) 
 
| 㡢ǿЩ IMNSearch (NetQuestion) ǵ
| ƫҡ桢ǵȴֱϯ ID ů ID
| ꢡҡ (ڦƪ IMNSearch )ģ̇
|  ID 㡤̧ĶȴԷϯȴ̽⡢
| ɳǡ ID ůϯ ID̥
| ҳ IMNSearch ǵȩ
 
| ġȹĸ IMNSEARCH 
 
| Ϊ IMNSearch ǵ̯ĴȴЩġ
| ܨ¡ƪǵѥҡǿƪ
| "imnadm" ϯ ID ů IDҡ
 
| ǿЩƶġԶ IMNSEARCH ǵĶ IMNSEARCH 롢
| ̯
 
| Ϊ IMNSearch ǵ̯ħɱ
|  IMNSearch ǵ (ƪ)Ŧƪܢȴ IMNSearch
| ǵ
Ű桢ȴġԶҳ imnadm  IDΪ
| ȴ㡢ƫ表
 
| 1.  ǡ imnadm ϯ IDѥ IMNSearch ǵ
 
| 2.  ɴ imnadm  IDѥ IMNSearch ǵ
 
|  AFS/DCE  NIS/NIS+  IMNSERACH ̯
 
| ܡϯ ID ů ID Ķ (ľ멡ܡ
| ϯΤǡ IDѥš ID ع̯ڷ׻)ϯ
| ǿǡ "imnadm"ϯ ID ů IDѥ IMNSearch
| ǡǵҳ IMNSearch ǵȴܨϯ ID
| ů IDǿԶǵġԶܨ¡
|  IMNSearch ǵ޲ѥġԶܨѥЬƪ
| ϯ ID ů ID
 
 
| 21.2   HPUX-10 ĸ NETQUESTION
 
|  DB2 Universal Database  5.2 for HP-UX 10 롢ƫ
| ܢ NetQuestion ڷܨ CD-ROM ҡܢ DB2
| ǵ롢ȭ̯ġЬ "Τ" عɡ
| Ķ NetQuestion
 
|      swlist -l product | grep -i imn
 
| ΪȴǶعɡЬ NetQuestion
 
| ĶٯȢ NetQuestion
| ٯ NetQuestion
 
| Զ "/NetQ" ڷ CD-ROM ̯㡤
| ȩѥЬ "/NetQ/hpux10" ȴĶ
| "IMNSearch""IMNSearchB""IMNpkgadd""IMNpkgaddB""imnbinstall"
| ů "imninstall"ī
 
|      imninstall
 
| ٯ NetQuestion  scriptĸƷ DBCS 
| ſ㡢 SBCS ů DBCS ſ㡤
 
| 
 
| Ϊ "imninstall"  "swcopy ٶ swcopy ٯ"
| 먡Ьƫ DB2 ̧˷˷ӡ
| ԶĶ (ٯ) Ķ̯ġԶܢ
| "swcopy""swremove"  "swinstall" "/NetQ/hpux10/IMNpkgadd"
| ʡ
 
|      -x mount_all_filesystems=false
 
|
 
 
| 21.3  UNIX ƺ  NETQUESTION ů DB2 CAE
 
| ĸǵ NetQuestion ȴ UNIX ܨ CD (Ʒȴ Client
| Application Enabler) ĸԡΪڷ EE  EEE ܨ CD  CAE
| Ьľƫĸǵǵ롢ٯ NetQuestion
 
 
| 21.4  AIX ƺĸ NETQUESTION DBCS
 
| DB2 Universal Database  5.2 for AIX  NetQuestion 
|  (DBCS) ſ㡢Ϊ̧ȹ NetQuestion
|  AIX ɢ˷ĸЬڦɢ
 
| Ϊơơ DB2 UDB V5.2 
| ȴ VisualAge C++ Professional  4.0 for AIX  DBCS
| ǵǶ AIX ɢ˷ĸΪȴ DBCS 
|  AIX 4.3 ɢ˷ĸЬ NetQuestionǶ
|  DB2 DBCS ǵԶҳ̧ NetQuestion ܨ
|  CGI  DB2 UDB  5.2 for AIX  CGI 
| թ
 
| Զ롢Ķ CGI 
| "/var/docsearch/cgi-bin/" ů "/usr/docsearch/cgi-bin/" 
|  CGi
 
| 1.  ڷĶĶ CGI (ҳ "db2srdbcs") ̯
      "/var/docsearch/cgi-bin" ů "/usr/docsearch/cgi-bin"
 
|          ftp.software.ibm.com/ps/products/db2/info
 
|     ܢ "db2srdbcs" 
 
| 2.  ůٯ DB2 DBCS ſ㡢Զ桢Ķ
 
|     o   ſ롢ī
 
|              usr/IMNSearch/bin imqss -stop dbcshelp.
 
|     o   ٯſ롢ī
 
|              usr/IMNSearch/bin imqss -start dbcshelp.
 
 
  22.0  SCOHELP ع
 
 
  ħĶ SCOhelp ع
 
  o   22.1, "̽ SCOhelp"
  o   22.2, "ե SCOhelp"
  o   22.3, "Web ̱ SCOhelp".
  o   22.4, "Support Level Supplement (SLS) PTF7004A, the UnixWare 7 Language
      Extension Supplement".
 
 
  22.1  ̽ SCOHELP
 
  ƫɳĶġȢ̽ SCOhelp
 
  o   ڷ硨 "scohelp" (/usr/bin/X11/scohelp)

 
  o   ڷ CDE ֤ĸܡȢ ""  (̱ƺ
      ĸ)ѥ̽ SCO 멡
 
  o   ڷ Panorama ֤ĸܡȢ "Root" Сѥ̽
      멡ס
 
 
  22.2  ե SCOHELP
 
 
  22.2.1  ֱ
 
  ̧㡢"UnixWare 7 Update 1 (7.0.1)" CD-ROM ħ
  (esle) ԺȢ Ķ DB2 ʩ LANG ҳ "es" 
  ɢħ UnixWare 7 Update 1
 
  o   ǡ (ġ DAS )ҳħɭԶǡ
      Щǿ LANG ҳ "en"  "C"ѥǡ롢
       "es" ľΪ LANG ҳ "es" 㡢
       DB2 Ȣ (db2setup) ǡ
 
  o    db2ishut ϯҳħɭԶ
      DB2ishut Щǿ LANG ҳ "en"  "C"ѥȩ롢
       "es"
 
  o    db2istrt: Ⱥٯ DB2 Τ¡
      /etc/rc.db2 script ƵǡġԶ롢Ķġ
      ɢ׻ٯ̯ /etc/inittab 㡢ٯ
      ٯ DB2 Τ٭Ȣ
 
           db:234:once:/etc/rc.db2 > /dev/console 2>&1 # Autostart DB2 Services
 
  Ϊ LANG ԫҳ "es" /etc/inittab ġ
  ҳެ⡡ٯȩ롢 root ǹīѥ LANG 
  ҳ "en"  "C" /etc/rc.db2 DB2 ٯѥ
  LANG  "es"
 
   UnixWare 7 Update 1 Ժߡ
  롢ּ¡ etc/inittab ּ̽ġ
  ެ (uncomment out) DB2 Τ٭ȢĶġȹٯ
  ٯٯ
 
       db:234:once:/etc/rc.db2 > /dev/console 2>&1 # Autostart DB2 Services
 
 
  22.2.2  Τĸ SCOHELP 
 
  ΪΤĸ SCOhelp ԫơ
  /usr/ns-home/httpd-scohelphttp/config/lang.conf
  DocumentRoot ҳơ
 
       DocumentRoot /usr/lib/scohelp/fr
 
  ġաĶٯ web Τ¡
 
       /usr/ns-home/httpd-scohelphttp/restart
 
  Ϊ桢Ь web Τ¡ѥĶٯơ

 
       /usr/ns-home/httpd-scohelphttp/stop
       /usr/ns-home/httpd-scohelphttp/start
 
 
  22.2.3   SCOHELP ڷ׻
 
  ڷ׻ƫ SCOhelp ع
 
  SCOhelp Τ *httpAcceptLanguage 㡢ȭ
  ġơƫĶȢ *httpAcceptLanguage
  ȩơ
 
  o   Netscape 3 Netscape app-defaults  (ġ
      /usr/X/lib/app-defaults) īȺ硨
 
           *httpAcceptLanguage: fr, en
 
  o   Netscape 4̽ҡѥ١ס
      ڷ̽ſ¡סѥ١ơס
      ơġԶ

 
  Ϊɻ *httpAcceptLanguage SCOhelp 㡢
  ڷƪơ
 
 
  22.2.4  
 
  Żȴ䡤
 
  ƫѺġĶ SCOhelp ĸѺڡ̽
  ơ̽ڡ
 
 
  22.3  WEB ̱ SCOHELP
 
  Web ̱ DB2 ȴ ( SCOhelp )
   22.1 ̽ SCOhelpġ SCOhelp
  ů̽ DB2 
 
  SCO DB2 Web ̱ȴ䡤
 
 
  22.4  SUPPORT LEVEL SUPPLEMENT (SLS) PTF7004A, UNIXWARE 7 LANGUAGE
  EXTENSION SUPPLEMENT
 
  "Support Level Supplement (SLS) PTF7004A" ƫڷĶ SCO FTP
  ׻ƺ̽ڵ
  ftp://ftp.sco.com/SLS/ Download the files ptf7004a.Z and
  ptf7004a.txt
 
  PTF7004A "SCOhelp  Verity ſů
  Ȣ SCOadmin ܡȢŧ" Ϊ DB2 ǵ
  ЬȺ SLSعٶ ptf7004a.txt
 
 
  22.4.1  ԺŻ LANG.CONF
 
  ġħ PTF7004AŻ SCOhelp
  Щǿ /usr/ns-home/httpd-scohelphttp/config/lang.conf
  Ķա

 
  1.  Ժ LocalizedDocRoot Ķ̥硢 "_BR" 
      "_CN" ̯ȹʢĶ
 
           LocalizedDocRoot pt_PT.ISO8859-1 /usr/lib/scohelp/pt_BR
           LocalizedDocRoot zh_ZH.ISO8859-1 /usr/lib/scohelp/zh_CN
 
  2.  Ժ zh ɱء zh ɱȴ̯
      zh_ZH.ISO8859-1 (zh  ZH ͵ע)
      թĶ
 
            LanguageAlias zh zh_ZH.ISO8859-1
 
      ȺԫȺ硨
 
            LocalizedDocRoot zh_ZH.ISO8859-1 /usr/lib/scohelp/zh_CN
 
  ħѥ 170 ĸ 22.2.2, "Τĸ
  SCOhelp ơݡٯ (ѥٯ) web Τ¡
 
 
  22.5  SCOHELP ſ
 
  SCOhelp ȴſ
  /usr/lib/scohelp/<lang>/_SearchIndex/parts 㡤
 
  ܨˡƪ
  ܨ (ܢ UnixWare 7 Ȣ)
  DB2 ̧ܨ¡
 
  ΪɴЬ SCOhelp 桤
 
 
