Dieter W. Storr Efficient Usage of Adabas Replication
Dieter W. Storr
Efficient Usage of Adabas Replication A Practical Solution Finder Edited by Stephen Fedtke
VIEWEG+ TEUBNER
Bibliographie information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographie data are available in the Internet at http://dnb.d-nb.de.
1st Edition 2011 All rights reserved
© Vieweg+ Teubner Verlag I Springer Fachmedien Wiesbaden GmbH 2011 Editorial Office: Christel Roß I Maren Mithöfer Vieweg+ Teubner Verlag is a brand of Springer Fachmedien. Springer Fachmedien is part of Springer Science+Business Media. www.viewegteubner.de No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permiSSion of the copyright holder. Registered andjor industrial names, trade names, trade descriptions etc. cited in this publication are part of the law for trade-mark protection and may not be used free in any form or by any means even if this is not specifically marked. Cover design: KünkelLopka Medienentwicklung, Heidelberg Printing company: STRAUSS GMBH, Mörlenbach Printed on acid-free paper Printed in Germany ISBN 97B-3-834B- 1730-3
Preface When I presented my experiences with the Event Replicator for Adabas at the International Users' Group Conference in Prague, 2010, Software AG and conference attendees inspired and encouraged me to write a book about replication to share my knowledge. This book will discuss the basies of the Event Replicator and show how to set-up files for replication and create the definitions for destination, subscription, GFB, initial state and the new ADADBS fuoctian.
Not a11 utility functions can currently be automatically replicated on a nonmainframe platform. An example is file changes that require special handling on the subscription, the replicator eogine and the target site. My book will walk you through the process step by step. From a disaster and recovery standpoint, copying databases to the same or other h ardware platforms becomes increasingly important. This book explains the difference between mirroring and replicating and shows the pros and cons of both tedmiques. In tests with production-sized data of 160 m illion records, communication problems occurred, parameters weren't calculated correctly and RPL buffers and SLOG overflowed. The initial-state procedure was too time-consuming and a workaround had to be found.
In production with 1,400 updates per second, the source database showed higher CPU time and batch jobs ran much longer when replication was turned on. Despite users' statements that the replication works correctly and comparisons between source and target are not necessary, seH-written comparison programs sometimes show differences. Three methods will be discussed how to compare source and target files to ensure that both are identical. Since database data is available in local or wide area networks and on public Web sites, security issues become even more important. One chapter in this book explains the available security software protecting the components during the replication process. Last but not least, this book will also show the very important recovery function. Normally, the Even t Replicator Server restarts automatically after an abn ormal end and is able to recover any lost replication. But in some cases, data is lost and the replication must be replayed.
VI
Preface
My special gratitude goes out to Larry Frazin, who supported me with replication tests. John Donnelly, Albert Stetson, Victor Tisuela, Hazel Baeza, Gary Walker and Chris Jonas who sent parameter values from their organizations. Roseanna Torretto monitored resources on the mainframe with RMF and Strobe. The team from Software AG, Wolfgang Weiss, Becky Albin, Alex Burggraf and WH Heynen, who were always available for tuning discussions and product sup-
port. My thanks also to Michelle McGowan for her proofreading skills and emotional support. Dieter W. Starr Sacramento, Califomia, U.S.A, August 2011
Contents Preface .. .... .... ... .. .. .. .. .... .... ... .. .. .... ... .. .. .. .. ... .. .. .. ... .. .. .. .... ... .. .. .... .... ... .. .. .... .... ... .. .. .. .. .... ... V Table of Figures ......................................................................................................... XIII
1
Replication Basics .............................................................................................. 1.1 Today's Challenge .................................................................................. 1.2 Replicating or Mirroring .. .. .. .. ..... .. .. ..... .. .. .. ............. .. ............. .. ............. 1.2.1 Mirroring ................................................................................................. 1.2.2 Replication .. ........... .............................. ........... ............... ........... ............... 1.3 Reasons for Replication ......................................................................... 1.4 Check Your Intentions Before Starting .......................... ...................... 1.4.1 Mirroring or Replication? ...................................................................... 1.4.2 Usage of Replicated Database .............................................................. 1.4.3 Measure/Compute and Determine Your Needs ................................ 1.4.4 Avoid Mistakes .......................................................................................
1 1 3 3 4 5 7 7 8 9 10
2
Different Ways to Replicate ............................................................................. 2.1 General Requirements I Basics ... .. .. .. ... .. .. .. .. ........... .. .. ........... .. .. ........... 2.2 From Adabas to Adabas on the Mainframe ....................................... 2.3 From Adabas on the Mainframe to Adabas on Linux, UNIX or Windows (LUW) ................................................................................ 2.4 From Adabas on the Mainframe to Non-Adabas Databases ...........
13 13 14 15 17
Definitions and Parameters ............................................................................. 3.1 Getting Started .......... .......................... ........... ............... .......................... 3.2 Source or Subscription Database (V8.1.3) ........................................... 3.2.1 Workload Manager (WLM) .............. ............... ............... ........... ........... 3.2.2 ADARUN Parameters ........................................................................... 3.2.3 ADADBS Replicate Files .. .. .. ... .. .. .. .. .. ... .. .. .. .. ........... .. .. ........... .. .. ........... 3.3 Replicator Engine - Reptor (V8.1.3N3.2.1) ........................................ 3.3.1 Workload Manager (WLM) ...... .. .. .. ........................ .. .. .. .................... .. .. 3.3.2 SLOC File ................................................................................................ 3.3.3 Replication System File .. ................... ............... ............... ...................... 3.3.4 ADARUN parameters ........................................................................... 3.3.5 Event Replicator lnitialization Parameters ......................................... 3.3.6 Define the Global Format Buffer .......................................................... 3.3.7 Define the Destinations .............. ....... ............... ............... ...................... 3.3.8 Define the Subscriptions ....................................................................... 3.3.9 Define the Transaction Filter Definitions ............................................ 3.3.10 Define the Initial-State ...........................................................................
19 19 22 23 23 24 25 25 26 26 27 29 31 35 36 37 38
3
Conlenls
VIII
Unload and load parameters ................................................................ Entire Net-Work (Wep V6.2.1) - Mainframe ..................................... AOARUN parameters ........................................................................ ... Parameter Statements ............................................................................ Entire Net-Work (WCP V7.3.3) - LUW .............. .. .. .. ..... .. ............. .. ..... Parameters ............................................................................................... Service.config ... .. .. .. .. ..... .. .. .. ..... .. ............. .. .. .. ......... .. .. .. ......... .. ............. ... KZOO8.KERNEL ...................................................................................... Xts.config ... ............. .. ..... .. .. .. ..... .. ............. .. .. .. ......... .. .. .. ......... .. ............. ... Entire Net-Work (WCL V1.3.1) CIient ................................................. Windows Service ................................................................................. ... UNIX Oaemon ........................................................................................ Xts.config .............................................................................................. ... Service.config .......................................................................................... Oirectory Server - xtsurl .cfg ................................................................. Adabas on Windows (V6.2.1.1) ............................................................ General ..................................................................................................... Parameters ............................................................................................... EntireX Broker (V8.0.1) .......................................................................... EntireX Broker (V7.3.4) .......................................................................... Event Replicator Target Adapter ......................................................... Schema and Replicated Oata ................................................................
38 41 41 42 43 43 43 44 44 44 44 45 45 45 46 46 46 46 48 48 52 53
4
Compare Files of Source and Target ... .. ............. .. ......... .. ..... .. .. .. ......... .. .. .. ..... 4.1 Compare Amount Fields ....................................................................... 4.2 API USR40]]N - User Exit ................................................................ ... 4.3 Compare Highest ISNs (fOPISN) .......................................................
57 58 59 60
5
Recovery After a Crash ..................................................................................... 5.1 Replay with a Token .............................................................................. 5.2 Replay Without a Token ........................................................................ 5.3 Replay-only Mode .................................................................................. 5.4 Initial-state .............. .. ......... .. .. .. .. ..... ......... .. .. .. ......... .. .. .. ..... .. ............. .. .....
63 64 67 68 68
6
Problems and Solutions .................................................................................... 6.1 Source or Subscription Oatabase .......................................................... 6.1.] Replicator Buffer (LRPL) Overflow ..................................................... 6.1.2 High CPU UsageofSource OB ............................................................ 6.1.3 AOAFRK - Transactions Tumed Off .................................................. 6.1.4 Source OB didn't come down ............................................................... 6.1.5 File Changes Won't Replicate ............................................................... 6.2 Replicator Engine (Reptor) ................................................................... 6.2.1 NAB Overflow ........................................................................................ 6.2.2 Replicator Buffer (LRPL) Overflow .....................................................
7] 71 71 72 79 79 80 86 86 87
3.' 3.5 3.5.1 3.5.2 3.6 3.6.1 3.6.2 3.6.3 3.6.' 3.7 3.7.1 3.7.2 3.7.3 3.7.4 3.7.5 3.8 3.8.1 3.8.2 3.9 3.9.1 3.9.2 3.9.3
Contents
6.2.3 6.2.4 6.25 6.2.6 6.2.7 6.2.8
6.2.9 6.2.10 6.2.11 6.2.12 6.2.13 6.2.14 6.2.15 6.2.16 6.2.17 6.2.18 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.35 6.3.6 6.4 6.4.1 6.4.2 6.4.3 6.4.4 6.45
6.4.6 6.4.7 6.4.8 7
IX
SLOG Overflow .... .... .... ....... .... .... .... ... .... .... ... .... .... .... ... .... .... ....... .... .... ... RSP 153 from the Target OB ................................................................ . RSP 113 from the Target OB .. .... .... ... .... .... ... .... .... .... ....... .... .... ....... .... ... RSP 98 from the Target OB .................................................................. . RSP 162 from theTarget OB .. .. ... .. .. .. ... .. .. .. .. .... .... ... .. .. .... .... ... .. .. .... .... ... Special characters are not correctly replicated .................................. . Wrong Replicated File in SLOG ... .. .. .... ... .. .. .. .. .... .... ... .. .. .... ... .. .. .... .... ... RSP 148 From Target OB ...................................................................... . Node is unreachable ... .. .. .. ... .. .. .. .. .. .. .. ... .. .. .. .. .... ... .. .. .... .... ... .. .. .... .... ... .. .. Communication Problem ..................................................................... . SLOG data disappeared ..... .... .... .... ....... .... .... ... .... .... .... ... .... ....... .... .... ... Replication is too slow .......................................................................... . Initial state is too slow ........ .... .... .... ... .... .... ... .... .... .... ... .... .... .... ... .... .... ... New AOAFRI Message ........................................................................ . Reptor under Stress .. .... .... ....... .... .... .... ... .... ... .... .... .... ... .... .... .... ... .... .... ... Replication with wrong FOT/GFB ...................................................... . Target Oatabase ..................................................................................... . RSP 52 from the Target OB .................................................................. . RSP 176 from the Target OB ................................................................ . RSP 98 followed by RSP 113 ................................................................ . RSP 113 on several files ........................................................................ . RSP 153 - Oisplayed on Reptor ..... ... .... .... ... .... .... ....... .... .... ... .... .... .... ... RSP 162 - LBP Space Problems ........................................................... . Utilities ..... ... .... .... .... ... .... .... .... ... .... .... .... ... .... ... .... .... .... ... .... .... .... ... .... .... ... Replay with more than 60 files ............................................................ . Replay with ERROR-139 .... .... .... .... ... .... .... ....... .... .... ... .... .... .... ... .... .... ... Replay with ERROR-140 RSP 254 ....................................................... . Replay with ERROR-140 RSP 131 Suhcode 70 ... .. .... .... ... .. .. .... .... ... .. .. Replay with ERROR-1 40 RSP 131 Subcode 61 .................................. . Replay with ERROR-I4$ - Token .. .. .... ... .. .. .. .. .... ... .. .. .. .. ... .. .. .... .... ... .. .. Replay Abended with SOC4 ................................................................. . Replay with Res ponse 131 Subcode 61 .. .... .... ... .. .. .. .. .... .... ... .. .. .. .. .... ...
Monitors .. .... .... .... .... ... .... ... .... .... .... ... .... .... .... ... .... .... .... ... .... .... .... ... .... ... .... .... .... ... 7.1 What can be monitored? ....................................................................... 7.1.1 Source OB on the Mainframe ................................................................ 7.1.2 Replicator on the Mainframe ..... .... ... .... .... ... .... .... .... ....... .... ....... .... .... ... 7.1.3 Entire Net-Work (WCP) on the Mainframe ........................................ 7.1.4 Network -LineorCircuit .. .... .... .... ... .... .... ... .... .... .... ....... .... ....... .... .... ... 7.1.5 Firewalls, Routers, Load Balancing Systems ...................................... 7.1.6 Entire Net-Work (WCP) on Windows ..... .... ... .... .... .... ... .... ... .... .... .... ... 7.1.7 Adabas on Windows - Command Log ...............................................
89 90 92 94 95 95 96 97
100 100 101 101 102 103 104 105 105 105 106 106 107 108
109 112 112 112 113 113 114 114 115 117 119 119 121 121 124 126 127 128 129
Conlents
X
Self-Written ........ ........... .................................. ............... ........... ........... ... RPL and NAß Buffer Overflow ............................................................ 5tOG check on Reptor .... ............... ........... ........... ............... ............... ... Compare two SLOGs and Open Destination ..................................... SLOG Check for Logged Files ... ........... .. .. .. .. ....... .. .. .. .. ....... .. .. ........... .. . StOG Space and Number of Records Check ..................................... Check Messages on DDPRINT .. .... .... ... .. .. .... ... .. .. .. .. .... ... .. .. .. .. .... .... ... .. . Target Check - Ping ............................................................................... Destination Report ........... .. .. .. .. ... .. .. ........... ....... .. .. .. .. ....... .. .. .. .. ... .. .. .. .. ... Subseription Report ............................................................................... RPL Flag Cheek in ADAREP .................... ............... ............... ........... .. . Analyze WCP Log .................................................................................. Determine Highest ISN Currently Used ................ ............... ........... .. . Compare File Components After Replieation .................................... Optimize for Infrastrueture (041) ................ ............... ........... ........... .. . General ..................................................................................................... Dimensions .............................................................................................. KPIs of Event Map AdabasEventRep .................................................. BuHt-In Rules .......................................................................................... Dimensions - Destination ..................................................................... KPIs of Event Map AdabasEventRepDest .......................................... BuHt-in Rules- Destination ..................................... .......................... .. . Dimensions - Input Queue ................................................................... KPIs of Event Map AdabasEventRepinQueue .................... ........... .. . BuHt-In Rules .......................................................................................... Dimensions - Subseription .... ....................... ............... ........... ........... .. . KPIs of Event Map AdabasEventRepSubscr ...................................... BuHt-In Rules ... .. .. .. .. .. ....... .. .. .. .. ... .. .. ........... ....... .. .. .. .. ....... .. .. .. .. ... .. .. .. .. ... How to Monitor a Component .............................................................
129 129 132 134 138 139 141 142 144 146 148 151 152 152 153 153 155 155 156 156 156 157 157 157 158 158 158 158 159
Security ................................................................................................................ Souree Database Adabas ....................................................................... Adabas SAF Security (ADASAF) ......................................................... Adabas Security (ADASCR) ................................................................. Adabas Data Eneryption (Ciphering) .................................................. Adabas Multi-elient Files ...................................................................... Replicator Adabas ............ .......................... ............... ............... ........... ... Adabas SAF Security (ADASAF) ......................................................... Entire Net-Work (Wep) Mainframe V6.2.1 .... ... .... .... .... .... ... .... .... ... .. . Entire Net-Work SAF Security (NETSAF) .......................................... Encryption Vl.1.2 .. .... ....... .... .... .... ... .... .... .... ... .... .... .... ... .... .... ... .... .... ... ... EntireX Broker ........................................................................................ EntireX Security ... .. .. .. ....... .. .. ... .. .. .. .. ....... .. .. .. .. ....... .. .. .. .. ....... .. .. ... .. .. .. .. ...
163 163 163 164 164 165 165 165 165 165 166 166 166
7.2 7.2.1 7.2.2 7.2.3 7.2.4 7.2.5 7.2.6 7.2.7 7.2.8 7.2.9 7.2.10 7.2.11 7.2.12 7.2.13 7.3 7.3.1 7.3.2 7.3.3 7.3.4 7.3.5 7.3.6 7.3.7 7.3.8 7.3.9 7.3.10 7.3.11 7.3.12 7.3.13 7.3.14
8
8.1 8.1.1 8.1.2 8.1.3 8.1.4 8.2 8.2.1 8.3 8.3.1 8.3.2 8.4 8.4.1
Contents 8.5 8.6 8.6.1 8.7 8.8 8.9 8.10 8.10.1 8.10.2 8.11 9
XI
Network .. ................... ........... ............... .......................... ............... ........... Entire Net-Work (WCP) LUW V7.3.4 .................................................. Encryption V1.1.2 .................... ............... ....... ............... ............... ........... Entire Net-Work (WCL) Client V1.3.3 ................................................. Event Replicator Target Adapter ... .. ... .. .. .. .. ........... .. .. ....... .. .. ........... .. .. Adabas ADASCR on Linux, Unix, Windows ..................................... Natural Security ........... ....... .. .. ... .. .. .. .. ....... .. .. .. .. ....... .. .. .. .. ....... .. .. ........... NSC .......................................................................................................... Natural SAF Security .......... .. .. .. .. .. ... .. ... .. .. .. .. ........... .. .. ....... .. .. ........... .. .. Non-Adabas on Linux, Unix, Windows .............................................
166 167 167 167 167 167 168 168 168 168
Append ix ............................................................................................................. 9.1 Subscription DB - ADARUN parameters ........................................... 9.2 Reptor -ADARUNparameters ...................... ............... ...................... 9.3 Definitions Target DB - Windows ....................................................... 9.3.1 Parameters - db251.ini .................. .............................. ............... ........... 9.3.2 High Water Marks .................................................................................. 9.3.3 Activities .................... .............................. ....... ............... ............... ........... 9.3.4 Buffer Pool Statistics .............................................................................. 9.4 Test Results .. .. .. .. ....... .. .. ....... .. .. .. .. ... .. .. .. .. ... .. .. ....... .. .. .. .. ........... .. .. ........... Natural Monitor Programs ................................................................... 9.5 9.5.1 Ping Target Database .. .. .. .. .. .. .. .. .. ... .. .. ... .. .. .. .. ........... .. .. ....... .. .. ........... .. .. 9.5.2 Check Messages on DDPRINT ............................................................. 9.5.3 RPL and NAB Buffer Overflow .. .. .. .. ... .. .. .. .. ........... .. .. ........... .. .. .. .. ....... 9.5.4 SLOG Check on Reptor ......................................................................... 9.5.5 Compare two SLOGs ....... ... .............................. ............... ........... ........... 9.5.6 Analyze WCP Log .................................................................................. 9.5.7 Checksum Program ............ .............................. ............... ........... ........... 9.5.8 Compare Amount Fields ....................................................................... 9.5.9 Parse Adabas Report for RPL Flag ...................................................... 9.5.10 Determine Highest ISN Currently Used ............................................. 9.5.11 RPLJOREP - SLOG Check ..................................................................... 9.5.12 READF30P - Read SLOG Content ....................................................... 9.5.13 RPL-POO2 - Compare File Components .............................................. 9.6 WCP Log Windows ............................................................................... 9.7 ADARPL Replay SOC4 Dump .............................................................. 9.8 Assembler Programs ........... .... .......................... ............... ........... ........... 9.8.1 Sieep/Wait ............................................................................................... 9.8.2 Get Dataset Name ............... .............................. ............... ........... ........... 9.9 EntireX V8.0.1 Parameters .....................................................................
169 169 171 173 173 175 175 176 177 180 180 181 183 185 186 189 190 192 194 197 200 203 204 206 213 216 216 218 220
Bibliography ............................................................................................................... 223 Index .................... ....... ............... ............... .............................. ........... ............... ........... 225
Table of Figures Figure 1: Figure 2: Figure 3: Figure 4: Figure 5: Figure 6: Figure 7: Figure 8: Figure 9: Figure 10: Figure 11: Figure 12: Figure 13: Figure 14: Figure 15: Figure 16: Figure 17: Figure 18: Figure 19: Figure 20: Figure 21: Figure 22: Figure 23: Figure 24: Figure 25: Figure 26: Figure 27: Figure 28: Figure 29: Figure 30: Figure 31: Figure 32: Figure 33: Figure 34: Figure 35: Figure 36:
Architecture of Adabas Replication on the Mainframe Source: Software AG ..... .. .. .. .. .. .. .. ..... .. .. .. ............. .. ............. .. ............. Mirroring - ADADBS TRANSACTrON SUSPEND Function .... Reasons for Replication .................................................................... Homemade Replication .... .. .. .. .. .. .. .... ... .. .. .. .. .... ... .. .. .. .. ... .. .. .... .... ... .. .. General Requirements - Buffer Usage ........................................... Replication from Adabas to Adabas on the Mainframe .............. Replication from Adabas to Adabas on Linux, UNIX or Windows (LUW) .......................................................................... Replication from Adabas to non-Adabas Databases ................... Determine Length of RBL and FBL .... ... .. .. .... ... .. .. .. .. ... .. .. .... .... ... .. .. Determine Number of Commands ................................................. Determine Type of Commands ....................................................... Compu tation for Mbps Bandwidth .... .... ... .... .... .... ... .... ....... .... .... ... Used Software Versions ................................................................... Buffer Usage in Subscription Database .......................................... Replication Status in Adabas Report .. ... .... .... .... ... .... .... .... ... .... .... ... Buffer Usage in Reptor ..................................................................... SLOG File Shown in Adabas Report .. ... .... .... .... ... .... .... .... ....... .... ... ADARUN Parameters of Reptor ..... .... .... ... .... .... .... ... .... ... .... .... .... ... Main Menu Adabas Replicator Subsystem ................................... Display Global Format Buffers Using Replicator Subsystem .. ... Add Global Format Buffer Using Replicator Subsystem ............ List and Select Predict Files Using Replicator Subsystem ........... Select a Predict Files Using Replicator Subsystem ....................... Select a Predict File Using Replicator Subsystem ......................... Save the Selected Predict File Using Replicator Subsystem ....... List Included New Predict File Using Replicator Subsystem ... .. Create New Destination Using Replicator Subsystem ................ Define New Destination Using Replicator Subsystem ................ Define New Subscription Using Replicator Subsystem ... .... .... ... Define New Transaction Filters Using Replicator Subsystem .... Perform Initial-State Using Event Replicator Subsystem ............ Destination Parameters .... .. .. .. .. .. .. .... ... .. .. .. .. .... ... .. .. .. .. ... .. .. .... .... ... .. .. Subscription Parameters .................................................................. Global Formats (GFB) from Predict ................................................ SYSRPTR Unload all Parameters in Batch .... ....... .... .... ....... .... .... ... SYSRPTR Load all Parameters in Satch .........................................
2 4 6 6 13 14
16 17 20 20 21 21 22 22 24 25 26 28 31 31 32 32 33 33 34 34 35 36 36 37 38 39 39 40 40 41
XIV
Figure 37: Figure 38: Figure 39: Figure 40: Figure 41: Figure 42: Figure 43: Figure 44: Figure 45: Figure 46: Figure 47: Figure 48: Figure 49: Figure 50: Figure 51: Figure 52: Figure 53: Figure 54: Figure 55: Figure 56: Figure 57: Figure 58: Figure 59: Figure 60: Figure 61: Figure 62: Figure 63: Figure 64: Figure 65: Figure 66: Figure 67: Figure 68: Figure 69: Figure 70: Figure 71: Figure 72: Figure 73:
Table of Figures
ADARUN Parameters (DDCARD) for Entire Net-Work (Wep) on Mainframe ....................................................................... Parameter Statements (DDKARTE) for Entire Net-Work (Wep) on Mainframe ....................................................................... service.config Entire Net-Work (Wep) on Open System ............ KZ008.KERNEL Entire Net-Work (WCP) on Open System .... ... xts.config Entire Net-Work (WCP) on Open System ................... xts.config Entire Net-Work Client (WCl) on Open System ....... service.config Entire Net-Work Client (WCl) on Open System xtsurl.cfg Directory Server Entire Net-Work Client (WCl) on Open System ................................................................................ Adabas Windows Parameters .. ....... .. .. ....... .. .. .. .. ....... .. .. .. .. ........... .. . Adabas Windows Buffer Pool Statistics ......................................... webMethods EntireX Parameters (V7.3.4) ..................................... Parameters for Target Adapter for SQl Database ....................... Generate Field Table - Source: SAG ............................................... Generated XML Schema 'Create' - Source: SAG .......................... Generated XMl Schema ' Insert' - Source: SAG ........................ .. . Generated XML Schema 'Update' - Source: SAG ........................ Generated XML Schema 'Delete' - Source: SAG .......................... Compare Source and Target Files - Amount Fields .................... Natural Program to Count Amount Fields ................................... Result of Natural Program to Count Amount Fields ................... Result of Natural Program to Count Amount Fields ................... Result of Natural Checksum Program ........................................... Compare Highest ISN (TOPISN) and Others .. ............................. Adabas Report Records loaded, TOPISN and MAXISN (Mainframe) ....................................................................................... Adabas Report Records Loaded, TOPISN and MAXISN (Windows) .......................................................................................... Compare Highest Used ISN OCl) .................................................. Compare Highest Used ISN (Output) ............................................ Replay Function and its Components ............................................ Replay Modes .................................................................................... Start Replay Vsing Event Replicator Subsystem .. ........................ Initiate Replay Using Event Replicator Subsystem ...................... ADARPL Replay Output without a Token ................................... Checksum Program to Compare After Replay .......................... ... SYSREPTOR - Perfonn Initial-State ............................................... SYSREPTOR - Perfonn Initial-State ............................................... lRPL Overflow ............... ....................... ............... ............... ........... .. . High CPU Time - 19:00 Start of 9 Batch Programs ......................
42 43
43 44 44 45 45 46 47 47 52 53 54 54 55 55 55 57 58 58 59 60 61 61 61 62 62 63 63 64 64 67 68 69 69 71 72
Table of Figures
xv
Figure 74: Figure 75:
73
Figure 76: Figure 77: Figure 78: Figure 79: Figure 80: Figure 81: Figure 82: Figure 83: Figure 84: Figure 85: Figure 86: Figure 87: Figure 88: Figure 89: Figure 90: Figure 91: Figure 92: Figure 93: Figure 94: Figure 95: Figure 96: Figure 97: Figure 98: Figure 99: Figure 100: Figure 101: Figure 102: Figure 103: Figure 104: Figure 105: Figure 106: Figure 107: Figure 108: Figure 109: Figure 110: Figure 111:
Duration of 9 Batch Jobs With and Without Replication ........... . Duration of 9 Jobs With and Without Replication and Source OB Parameters ..................................................................... . Duration of 9 Jobs Without and With Replication ...................... . Strohe Report - Adabas Source Database ..................................... . Strohe Report - Adabas Replicator Engine .. .... .... ... .... .... ....... .... ... Buffer Usage Between Source OB and Reptor ............................. . High CPU Time - Adabas Source Database - Used Parameters ADADBS REPLICATION .nd DSBI=OFF .... ... .. .. .... ... .. .. .. .. .... .... ... High CPU Time - Event Replicator Subsystem Global Parameters ............................................................................ . ADADBS REPLICATION DEACTIVATE ... .. .. .... .... ... .. .. .. .. .... .... ... ADAFRK - Transactions Turned Off ........................................... .. Source OB Waiting for Replication Message ................................ . Manual File Change Procedure - Dataflow .. .. .. .. .. .. ....... .. .. ......... .. Manual File Change Procedure - SLOG Check .......................... .. Manual File Change Procedure - SLOG Check .......................... .. Manual File Change Procedure - FTP Predict Changes to Windows ...................................................................................... .. Manual File Change Procedure - Predict Change Report ........ .. Manual File Change Procedure - Dataflow ................................ .. Manual File Change Procedure - Open the Destination ........... .. NAB Overflow on Reptor .............................................................. .. N AB Overflow Message ................................................................ .. NAB Overflow High Water Mark ................................................ .. LRPL Overflow Reptor .. .... ....... .... .... .... ... .... .... .... ....... .... .... ....... .... ... LRPL Overflow Reptor High Water Mark .................................. .. Adabas Report SLOC file ............................................................... .. Response Code 153 Shown in Reptor .. .. ....... .. .. .. .. ....... .. .. .. .. ....... .. .. Response Code 153 - CLOG Adabas Windows ......................... .. Response Code 153 - not shown in CLOC of Adabas Windows .. ... .. .. .. .. .. .. .. .. .... ... .. .. .. .. .... ... .. .. .. .. ... .. .. .... .... ... .. .. Response Code 113 - Message in Reptor ..................................... .. Response Code 113 - ISN Reusage ............................................... .. Response Code 98 Shown in Reptor and Parse Program .......... .. Response Code 162 Shown in Reptor ........................................... .. Response Code 162 - LBP and WRITE_LIMIT ........................... .. Special Character not Translated Correctly ................................ .. Refresh SLOG .................................................................................. .. Response Code 148 - Entire Net-Work or OB not Up ............... .. Response Code 148 -NAT3148 ...... .... .... ... .... .... .... ... .... ....... .... .... ... System Management Hub - Connectivity Check ....................... ..
73 74 74 74 75 76 77
78 79 79 79 80 81 81 82
83 85 86 86 86 87 88 88 89 90 90
91 92 93
94 95 95
96 97 97 97 99
XVI
Figure 112: Figure 113: Figure 114: Figure 115:
Table of Figures
Computer Management - Services Check .... .... .... ...................... .. . 100 Node is Unreachable ........................................................................ 100 Communication Problem .. .... ................... .... ........... .... ....... .... ....... ... 100 ADAMACIC - Apply ADASA V from Mainframe Directly to Windows ........................................................................................ 103
Figure 116: Figure 117: Figure 118: Figure 119: Figure 120: Figure 121: Figure 122: Figure 123: Figure 124: Figure 125: Figure 126: Figure 127: Figure 128: Figure 129: Figure 130: Figure 131: Figure 132: Figure 133: Figure 134: Figure 135: Figure 136: Figure 137: Figure 138: Figure 139: Figure 140: Figure 141: Figure 142: Figure 143: Figure 144: Figure 145: Figure 146: Figure 147: Figure 148: Figure 149: Figure 150: Figure 151:
ADAFRI Message - SLOG Tumed Off .. ........... .......................... ... Reptor Under Stress - User Gone Message ................................... Response Code 55 - Wrong FDT/GFB ........................................... Response Code 52 Showed in Reptor ............ .. .. .. ......... .. ............. .. . Response Code 176 Showed in Reptor ........................................... Response Code 98 and 113 Showed in Reptor .............................. Response Code 113 Showed in Reptor from Target .. .. ............. .. . Response Code 153 Showed in Reptor - from WCP/WCL ......... Response Code 162 Showed in Reptor from Target DB .............. Response Code 162 - LBP and WRITE_LlMIT on Target DB ..... Response Code 162 - LBP and WRITE_LlMIT on Target DB ..... Response Code 162 - LBP and WRITE_LlMIT on Target DB ..... Response Code 162 - LBP and WRITE_LlM IT on Target DB ..... Response Code 162 - LBP and WRITE_LlMIT on Target DB ..... ADARPL Replay > 60 Files .............................................................. ADARPL Replay ERROR-139 ......................................................... ADARPL Replay ERROR-140 RSP 254 .......................................... ADARPL Replay ERROR-140 RSP 131 SC 70 ............................... ADARPL Version and Zap Level... .. ..................... .... ....... .............. ADARPL Replay ERROR-140 RSP 131 SC 61 ............................... ADARPL Replay -Create Token .................................................... ADARPL Replay ERROR-148 ......................................................... ADARPL Replay SOC4 ..................................................................... ADARPL Replay SOC4 - RPL Cleanup ......... .. .. .. ......... .. ..... .. .. .. ..... ADARPL Replay SOC4 - Create Multiple Token ......................... ADARPL Replay SOC4 - RPL Cleanup .......................................... Monitoring Replieation Parts - Graphie ........................................ Monitoring Replieation Parts - Tools ............................................. Monitoring - Number of Commands per Seeond ........................ Monitoring- High Water Marks- Souree DB .. .. .. .. .. .. .. ............. .. . Monitoring High Water Marks - Reptor ....................................... Monitoring TLOG - Adjustments ................................................... Monitoring TLOG - ADARPP ........................................................ Monitoring TLOC - Explaining Elements ..................................... Monitoring Destination Statistics on Reptor ................................. Monitoring Entire Net-Work (WCP) - Mainframe ......................
103 104 105 105 106 107 108 109 109 110 110 110 111 111 112 112 113 113 114 114 114 115 115 116 116 117 119 119 120 121 122 122 123 124 124 125
Table of Figures Figure 152: Figure 153: Figure 154: Figure 155: Figure 156: Figure 157: Figure 158: Figure 159: Figure 160: Figure 161: Figure 162: Figure 163: Figure 164: Figure 165: Figure 166: Figure 167: Figure 168: Figure 169: Figure 170: Figure 171: Figure 172: Figure 173: Figure 174: Figure 175: Figure 176: Figure 177: Figure 178: Figure 179: Figure 180: Figure 181: Figure 182: Figure 183: Figure 184: Figure 185: Figure 186: Figure 187:
Entire Net-Work (WCP) - MainframeDiagnostic Utilities GCL) ................................................................. Monitoring Network - NetQoS ............. ............... ............... ........... Monitoring Network - Browser, GUt iPhone .............................. Monitoring Firewall Analyzer - ManageEngine .......................... Entire Net-Work (WCP) on Windows Log .................................... Natural Parse Program for WCP on Windows Log ..................... Adabas Command Log on Windows ............................................. Check NAß and RPL Overflow GCL) ... .. ......... .. .. .. ......... .. ............. Check NAß and RPL Overflow (Output) ...................................... SLOG Check on Reptor GCL) .......................................................... SLOG Check on Reptor (Input) .......... .. .. .. ......... .. .. .. ..... .. ............. .. .. SLOG Check on Reptor (Output) .................................................... Compare two SLOGs Open Destination GCL) ............................. SLOG Check for Logged Files ... ......... .. .. .. ......... .. .. .. ..... .. ............. .. .. Manual File Change Procedure - SLOG Check ............................ Monitor Program: SLOG Check for Logged Files - READF30P SLOG Number of Records and Space Check GCL) ........... ........... SLOG Number of Records and Space Check (Output) ............... Read Output Queue of ActiveJob - EJESLNK GCL) ................... Output of Natural Parse Program .. ........... ..................................... wcr, Network and Target DB Check - r ING - GCL) ................ SYSAOS ßatch - Display Reptor StatisticsDestinations GCL) .......... .............................. ............... ........... ........... SYSAOS Batch - Display Reptor Statistics - Destinations Output ... .......................... .............................. ............... ........... ........... SYSAOS Batch - Select one Destina tion - Output ....................... SYSAOS ßatch - Display Reptor Statistics Subscriptions GCL) .. .. .. .. .. .. ... .. .. .. .. ....... .. .. .. .. ....... .. .. .. .. ... .. .. ........... .. .. SYSAOS Batch - Display Reptor StatisticsSubscriptions - Output ..................................................................... SYSAOS ßatch - Select one Subscription - Output ... .. ............. .. .. Check RPL Flag GCL) ....................................................................... Adabas Report RPL Flag Check - (Output) .................................. RPL Flag Check Report - (Output» .. .. .. .. ......... .. .. .. ..... .. ............. .. .. Analyze wcr Log ............................................................................. Create and FTP an Adabas Report to a Windows Server ........... Create Adabas Report and FTPtoaWindowsServer ................. Analyze wcr Log (Source: SAG) ................................................... Optimize for Infrastructure (041) - Analytics Overview ............ Optimize for lnfrastructure (041) - KPI Subscription User Transaction .........................................................................................
XVII
125 126 127 127 128 128 129 131 131 132 133 133 138 138 139 139 140 141 141 142 143 144 145 145 146 147 147 149 150 151 151 152 153 154 159 159
XVIII Figure 188: Figure 189: Figure 190: Figure 191: Figure 192: Figure 193: Figure 194: Figure 195: Figure 196: Figure 197: Figure 198: Figure 199: Figure 200:
Table of Figures Optimize for Infrastructure (041) - KP' Destination Replicated Transactions ................................................................... Optimize for lnfrastructure (041) - Destination Adabas Calls Committed ............................................................................... Optimize for Infrastructure (041) - Total Replicated Transactions .. ....... .... .... ........... .... ........... .... ........... .... ........... .... ....... ... Adabas SAF Security (ADASAF) - Source: SAG .......................... Adabas Multi-dient File ................................................................... WCP Encryption - Source SAG .......... .. .. ....... .. .. .. .. ....... .. .. ... .. .. .. .. ... ADARUN Parameters Subscription Database .............................. ADARUN Parameters Reptor ......................................................... Adabas for Windows Parameters db251.ini ... .. .. .. ....... .. .. ........... .. . Adabas for Windows High Water Marks ...................................... Adabas for Windows ADANUC Activities .................................. Adabas for Windows Buffer Pool Statistics ... .. ........... .. .. ........... .. . Replication Process - Number of Commands per Second
160 160 161 164 165 166 170 172 174 175 175 176
(cps) ..................................................................................................... 177
Figure 201: Figure 202: Figure 203: Figure 204: Figure 205: Figure 206: Figure 207: Figure 208: Figure 209: Figure 210: Figure 211: Figure 212: Figure 213: Figure 214: Figure 215: Figure 216: Figure 217: Figure 218: Figure 219: Figure 220:
Replication Process - Number of Commands per Second Spreadsheet 1 ..................................................................................... Replication Process - Number of Commands per Second Spreadsheet 2 ............... .... ....... .... ........... .... ........... .... ........... .... ....... ... Monitor - Natural Program Ping Target Database ...................... Monitor - Natural Program Check DDPRINT Messages ........... Monitor - Natural Program RPL and NAB Overflow... ........... .. . Monitor - Natural Program SLOG Check Reptor ........................ Monitor - Natural Program Compare 2 SLOGs .. ........... .... ....... ... Monitor - Natural Program Analyze WCP Log ........................... Monitor - Natural Program Checksum ......................................... Monitor - Natural Program Compare Amount Fields ................ Monitor - Natural Program Parse ADAREP for RPL Flag ......... Monitor - Natural Program ISNHIGH - Check Highest ISN .... Monitor - Natural Program RPL30REP - SLOG Space Check ... Monitor - Natural Program READF30P - Read SLOG Content Monitor - Natural Program RPL-P002 - Compare File Components .. ........... .. .. .. ... .. .. .. ........... .. .. ....... .. .. .. .. ....... .. .. .. .. ... .. .. .. .. ... Monitor - WCP Log Windows ........................................................ RPL SOC4 Dump ................................................................................ Assembler Program Sleep/Wait .............. .... ........... .... ....... .... ....... ... RPL SOC4 Dump ................................................................................ EntireX V8.01 Parameters ................................................................
178 179 180 183 184 186 189 189 191 193 196 199 202 203 206 212 215 217 219 222
1
Replication Basics
1.1
Today's Challenge
In today's world you must keep your enterprise competitive and your customers happy, therefore, it is critically important that your business systems have instant
access to changed data. Same organizations copy their Adabas datahase data to a data warehouse or to a relational database (RDBMS) located on Linux, UNIX or Windows, and use offthe-shelf software to process the copied data. Time and cast savings are also reasons to access copied Adabas data on nonmainframe platforms. Some batch jobs ruo on these platforms faster and cast less compared with running them on the mainframe. But it can be awkward and timeconsuming to decompress the data, file transfer it to the open system, campress and reload it.
Other organizations duplicate their Adabas database for disaster and recovery using backups and even protection logs to be up-to-date. This data-copy procedure is only possible on the same hardware platform. Open transactions on protection logs or handling problems with back-ups can easily lead to a corrupted target database. A third-party software company supports this approach with protection logs and even copies mainframe ADASA V back-ups to non-mainframe platforms. New Adabas releases on the mainframe and non-mainframe platforms makes it difficult or even impossible to keep up with Adabas format changes. The drawback of these methods is that your customers do not have instant access to the changed data, and the data is not updated in real-time. Another way is to mirror on a hardware-level your data or disks. Although synchronized mirroring is an update in real-time, you do not have instant access to the mirrored data. In addition, disk data errors will also be mirrored. Companies who want to copy parts of a file such as sets of records and fields or field values to the target database must write programs for this approach and then maintain this homemade software. The most practical solution for these problems is the Event Replicator for Adabas, a highly flexible event-publishing tool that efficiently deli vers changed data in real-time from Adabas to Adabas and to any RDBMS or middleware. Only committed Adabas modifications on a transaction level will be asynchronously replicated. D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_1, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
2
1 Replication Basics
Event Replication tor Adabas on Mainframe Architecture
,,
-,
SO"P, HTTP, SMTP, JOYO, .""',
""\lvo,
1
,.,............ ~,,~
I ,~
Intel,atlon PI.tfo<m
wM Entir ..X
IBM MQSeris
1= ' 1
Figure 1: Architecture of Adabas Replication on the Mainframe - Source: Software AG
The following chapters explain in detail the replication process and its parameters, show the architecture of various replication models, and give YOll guidance for possible problems and their solutions. One chapter is dedicated to security and describes how to protect each level in yOU f replication path. Examples of parameters, Natural and assembler source code complete this book as a guide and support for Adabas replication questions. • • • • • • • • • • •
The difference between replication and mirroring: chapter 1.2 The reasons for replication: chapter 1.3 What to check before starting and avoid mistakes: chapter 1.4 Different ways to replica te: chapter 2 Definitions and parameters: chapter 3 Compare files of source and target - self-written programs: chapter 4 Recovery after a crash: chapter 5 Problems and solutions based on own experiences: chapter 6 Monitors - self-written and from Software AG: chapter 7 Security: chapter 8 Appendix - many programs and parameter examples: chapter 9
1.2 Replicating or Mirroring
1.2
3
Replicating or Mirroring
For many years, companies backed up their mainframe databases to tapes and stored them at a very safe place, sometimes deep in a mountain storage. In case of a recovery after a disaster, the computer center mostly could not receive the backup tapes because streets, tunnels or bridges are unusable and airplanes are not aUowed to fly. In other cases the packages sent by the posta! service arrived only with a few tapes. One missing tape out of forty backup tapes is no backup at all. From a security stand point unacceptable to lose sensitive data, such as social security or credit card numbers.
1.2.1
Mirroring
One possibility to eliminate the above mentioned problem with tapes is to mirror the data from one disk to another disk. Mirroring is based on microcode on a disk array controller or via server software. It is typically a proprietary solution and not compatible between various storage vendors. Examples are EMC 5700 or enterprise storage server ESS Shark from IBM on enterprise servers (mainframes). Using peer to peer remote extended distance (PPRC-XD) from IBM, data will be copied over thousands of mHes between two ESS Sharks. SYMMETRIX remote data facility (SRDF) from EMC mirrors synchronized data to a remote back-up site, where the source and the target storage can be located on an enterprise server (mainframe) or on an UNIX box. Benefits
• Asynchronous disk mirroring can provide better physical protection by supporting extended physical distances. • No loss of committed transactions in synchronous storage (mirroring/RAID) on a CPU failure. Limitations • No protection from data corruption. • Secondary site is not guaranteed to be transitionaIIy consistent, in the case of asynchronous mirroring. • Client application must be re-started after faHure and need to be aware of failure. • Synchronous mirroring and RAID devices can add overhead to application performance. • Redundant/specialized high availability hardware/software can be expensive and restricted to use for backup purposes onIy. • Secondary copy of data is not available for use - Iow hardware utilization. • Need to replicate everything on disk, no seIectivity of data replication.
1 Replication Basics
4
IBM's FlashCopy copies nearly instantaneously entire volumes to other volumes (snapshot). It sets pointers and copies tater the entire image of volumes. The TimeFinder funcHon from EMC works similar. The miefar funcHon must be very carefully used in regards to Adahas and Hs buffer pooL Updated blocks might be still in the buffer pool or disk l/Os are only partially done when the fla sh copy function is used . The new ADADBSTRANSACfIONS-SUSPEND function takes care of the buffer flush and new update commands synchronizing it during the snapshot function. Without the transaction-suspend function, the database must be down during flash copy to avoid a datahase corruption. Pre-defln&d
timewindow Suspend
Read only: update requests are queued Resume
Read
~.J. Source
Data
Source: SAG
ADADBS TRANSACTIONS SUSPEND,nSYN=60,TRESUME=120
Figure 2:
1.2.2
Mirroring - ADADBS TRANSAcnON SUSPEND Function
Replication
Mirroring is very often called replication. But there are two main differences, Mirroring copies the tracks of a disk or 1/0 blocks and the mirrored data cannot be direct1y used for read-onIy processes. Replication repeats the database transactions on the target site and the replicated data can be instantly used for readonly processes. Benefits
• Warm Standby systems can be configured over a Wide Area Network, providing protection from site failures. • Ability to more quickly swap to the standby system in the event of failure, as backup database is already on-line. • Data cormption is typically not replicated as transactions are logically reproduced rather than VO blocks mirrored.
1.3 Reasons for Replication
5
• Automatie switeh over for clients using a switching mechanism, no dient restart needed. • Originating applications are minimally impacted as replication takes plaee asynehronously after commit of the origina ting transaction. • The warm standby database is available for read-only operations, allowing better utilization of backup systems. • Ability to resynehronize and easily switch back to primary system when it becomes available without loss of data. • No need to replicate everything on disk, data ean be selected for replieation. Limitations
• Warm Standby system will be out-of-date by transactions committed at the active database that have not been applied to the standby. • Protection is limited to components supporting Warm Standby (e. g. DBMS data sources may be protected but file systems may not be supported).
1.3
Reasons tor Replication
In today's finandal environment, many companies and state and federal agendes look at migrating their core applications from the mainframe, Natural and Adabas to newer hardware platforms, database systems and programming languages Iike Orade, SQL database, Unix and .NET or Java . However, after a cost analysis, they cannot always justify the expense or get permission from their stockholders to spend a couple of million dollars just to be more in the trend.
There is no proof that such a migration leads to faster processing or is more seeure or even easier to use. They also do not take into account the cost of losing all of the expertise in Natural and Adabas and retraining staff members just to be more hype. I can remember back in the 90s, when companies were so proud to own an adaptable database system and a 4GL programming language. Programs developed on the mainframe could easily run under other hardware platforms, as Windows, UNIX, or 652000. Some organizations found it is faster and less expensive to program in Natural under MS Windows and use Single Point of Development (SPoD) to coordinate and transport programs between the mainframe and the pc. They decompressed relevant Adabas files on the mainframe, transferred them via FTP to the PC, compressed and re-loaded them into the Adabas da tabase running under Windows. They also found that some batch jobs using Natural and Adabas ran on a pe up to 90% faster and cost less money than on the mainframe. Surprisingly, some other jobs ran 100% slower but this could be that Adabas Fastpath, the optimization tool of SAG, is not available on the pc.
1 Replication Basics
6
[
-
1 ASSO
DUA
WOIK
Figure 3:
Reasons for Replication
._.
This type of replication was homemade, very time consuming and left the cast of decompressing jobs on the mainframe site .
Cop,. ef Productlon
o .....s.
ASSO
OAYA
WORK
Figure 4:
Homemade Replication
The next challenge was how to populate the tatest Adabas data faster from the mainframe to the pe in order to be up-ta-date. Replication is the buzzword and some companies advertise products to copy Adabas data by using back-ups and protection logs. Software AG offered to help with their product "Event Replicator for Adabas" to replicate data in real-time from the mainframe to Linux, UNIX and Windows.
1.4 Check Your Intentions Before Starting
7
For many years, companies have used the Event Replicator for Adabas to replicate for disaster and recovery, development, training. demos and Web access. Off-the-shelf software for non-Adabas databases requires leaving the Adabas databases on the very secure and stable mainframe and replicate its data to nonAdabas databases on Linux, UNIX or Windows (LUW).
1.4
Check Your Intentions Before Starting
1.4.1
Mirroling 01 Replicalion?
Chapter 1.2 explains the two main differences between mirroring and replication. Mirroring is often called replication. The updated blocks in the Adabas buffer pool that have not yet been written to the disks can cause problems for mirroring. Adabas read commands receive the updated information from the buffer pool if it is still not written by a buffer flu sh to the disko In contrast to the replicated data the mirrored data is not up-to-date. It is disk mirroring and not buffer mirroring. The replicated target has the benefit that you have instant read-only access to the in real-time delivered data. Can you take advantage of this benefit for your organization? The following questions should help you in yOUf decision process. You also will find in the following chapters detailed explanations to the most important questions about Adabas data replication. • 00 you want to use you r copied data in paral1el to your original database? You can run batch reports against the replicated database and save money, time, and decrease mainframe workload: see chapter 1.3. If the answer is yes, you must use replication. • 00 you accept the possibility that errors in disk tracks or blocks are copied to the target? Mirroring might already be supported by your hardware (raid technology). If you have a micro-code problem on your source disk and would like to switch your production over to your mirrored target database this won't help because the error was mirrored . There might be an error recovery possible bu t it will take time. If you accept these errors then mirroring can be the solution. • 00 you want the Adabas transaction to be asynchronously repeated in near real-time on the target database? If you want to use your replicated data instantly then you must use replication.
8
1 Replication Basics
1.4.2
Usage 01 Replicated Database
It is important for yOll to decide what to do with your replicated Adabas database
after it is copied from the mainframe. Chapter 2 explains different ways to replicate • General requirements, basics: chapter 2.1. • From Adabas to Adabas on the mainframe: chapter 2.2. • From Adabas on the mainframe to Adabas on Linux, Unix, or Windows: chapter 2.3. • From Adabas on the mainframe to non-Adabas on Linux, Unix, or Windows: chapter 2.4. The following questions should help you in your decision process. •
•
•
•
•
00 you want to provide areal-time database for disaster and recovery and replicate your Adabas database from the mainframe to another Adabas database on the mainframe? Even if it is a replicated database for disaster and recovery, the data can be used instantly for read-only accesses. lf the answer is yes, you must use the Event Replicator of Adabas. See chapter 2.2. 00 you want to reduce the mainframe costs and replicate in real-time your Adabas database from the mainframe to Adabas on Linux, UNIX or Windows? lf the answer is yes, the Event Replicator of Adabas is your choice. See chapters 1.3 and 2.3. 00 you want to create or support a data warehouse with real-time data from your Adabas database on the mainframe? Software AG supports data warehouse appliance vendors and their database architectures, such as for 062, MS SQL Server, MySQL, Orade, Sybase, Sybase ASE, and Teradata. lf the answer is yes, you must use the Event Replicator of Adabas and its target adapters. See chapter 2.4. 00 you want to replicate your Adabas database from the mainframe to Linux, UNIX or Windows to a non-Adabas database so you can use yOUT off-the-shelf software, which is written for relational databases only? Software AG supports several target adapters, such as for 062, MS SQL Server, MySQL, Orade, Sybase, Sybase ASE, and Teradata. lf the answer is yes, you must use the Event Replicator of Adabas. See chapter 2.4. 00 you want to keep your sensitive data on the secure and stable mainframe and use YOUT existing message-oriented middleware such as MQSeries to copy in real-time alt or parts of the data to a RDBMS on the open system? lf the answer is yes, the Event Replicator for Adabas is the best solution. See chapter 2.4.
1.4 Check Your Intentions Before Starting
9
00 you need Entire Net-Work (WCP) in order to replicate Adabas data from mainframe to mainframe or to Linux, UNIX, or Windows? You must use Entire Net-Work as a broker funcHon. You cannot use webMethods EntireX. For me it makes more sense to use one broker function to replicate Adabas to Adabas and Adabas to non-Adabas databases. In contrast to Entire Net-Work, webMethods EntireX has Hs own reliable safety mechanism (persistent store). See chapter 2.2. • 00 you need a message-oriented middleware, such as MQSeries, or a broker function, such as webMethods EntireX? 00 you already use a broker function and Natural RPC getting mainframe Adabas data from a non-mainframe platform? See chapter 2.4. •
1.4.3
Measure/Compute and Determine Your Needs
Moving through the process on the way from the subscription or saurce database to the destination or target database must be passed, such as middleware, broker, network, router, and firewall . Parameters and the bandwidth of the circuit have to be detennined or adjusted. It is therefore a good idea to detennine your needs before you start with replication. The following questions will heIp to determine what information is important to collect and you wiJI find the answers in chapter 3. • How many Adabas modification commands (new, delete, update) do you expect per second? This information is important to detennine buffers in Entire Net-Work as weil as apart of the fonnula to determine the bandwidth (Mbps) to be used for the connectivity. See chapter 3.1. • How long are the record (RB) and format buffers (FB) per Adabas command? The number of commands per second is part of the formula to determine the bandwidth (Mbps) to be used for the connectivity. See chapter 3.1. • What line speed (circuit) and what type of connection do you need between the mainframe and Unux, UNIX or Windows servers, such as 0C3 or IPSec VPN tunnel? The line speed is crucial for the in real-time replication and shouId be determined for peak times. See chapter 3.1. • 00 existing lines have sufficient speed and are they shared with other applications, such as Web sites or ftp processes? If you calculate a line needed for 20 Mbps then an OC1 line with 51.8 Mbps can be too slow if used also by other applications. And keep in mind that the replication process sends back an acknowledgement to the Replication Engine. A duplex circuit would be a benefit. See chapter 3.1. • Are the Adabas modification commands send from batch and/or online programs? Batch programs send Adabas commands faster than online users can hit the enter-key. The number of Adahas commands is important to determine
1 Replication Basics
10
buffers in Entire Net-Work as weil as apart of the formu la to determine the bandwidth (Mbps) to be u sed for the connectivity. See chapter 3.t. • How many Adabas files should be replicated? The number of files per transaction is a factor in the formula to determine the number of attached buffers (NAß) of the Replication Engine. The Replicator Engine keeps all file updates of one transaction before sending them to the target database. See chapter •
3.3.4. 00 you plan to replicate only same fields of a file? The throughput of data rep-
lication to Adabas destinations is improved with version 3.3. SPl of the Event Replicator for Adabas by filtering format buffers. New parameters allow to specify before (SFFILTERGFBBI) and after-image (SFFILTERGFBAI) format buHers. This will reduce the decompress process time of the entire buHer. See chapter 3.3.9. • How many records per file? This information is important to determine the time needed for the initial-state process. Files with over 100 million records can take hours. The initial-state function reads record by record and sends store-with-ISN commands (N2) to the target database. Tests showed that based on an Adabas command log on Windows a fast connection can replicate approximately 300 Adabas commands per second. See test results in appendix, chapter 9.4. • 00 you need additional broker software, such as EntireX or Entire Net-Work?
1.4.4 Avaid Mistakes Ouring many years of experience with replication for Adabas V3.2.1, Iran into some problems and would like to help you to avoid the following mistakes. Problems and solutions are explained in chapter 6. •
• •
•
•
00 not create your replicated database on a pc, use a server instead. A local drive (C-drive) on a PC might not handle the traffic, use disk arrays and raid technology (RAID 5) in a SAN box to store the replicated database. 00 not log or trace your replicated data in a production environment. Oepending on the volume of replicated data, it can slow down the replication process. 00 not log the before image (OSBI=OFF) if it is not necessary. It is an overhead in data transfer between the subscription database and the Replication Engine. Chapter 3.2.3 shows the AOADBS commands. 00 not use slow lines (circuits) as a connection between the mainframe and open system servers (Linux, UNIX, Windows). A data traffic jam can affect the subscription database and the Replicator Engine, such as CPU time, buHer and SLOG overflow . See chapters 6.2.2 and 6.2.3 for more information . 00 not define your Adabas and replication buffers too smalL It can stop the replication process on the subscription database because there is no overflow area. Only the Replicator Engine has such an area (SLOG).
1.4 Check Your Intentions Before Starting
11
• Define your SLOG large enough to keep the replicated data for a couple of days if the target database is temporarily not available. lnitial-state and/or the REPLA Y function are necessary to restore the missed data. • 00 not underestimate the time you will need for the initial-state process, especially if files with large amount of records are involved. Attachment chapter 9.4 gives you an idea of commands per second (cps) going through all the replication components. • Keep in mind that one Replicator Engine may not be sufficient. The number of update commands per second, the number of files in one transaction, and the number of files to be replicated are responsible for the decision. • 00 not replicate files without a valid destination. It can be a big task to c1eanup the SLOG file during an ongoing replication process. See chapter 6.2.9. • 00 not wait to install monitor routines or software to check error messages or response codes reported on the subscription database and Replicator Engine (DDPRINT) - especially for ISNs not replicated. Chapters 4 and 7 explain how to compare and what monitor routines are available. Plan to have routines in place when file changes must be applied during a program migration. File change replications from Adabas to Adabas won't be done automatically in some cases. Chapter 6.1.5 explains in detail what to do.
2
Different Ways to Replicate
2.1
General Requirements I Basics
The subscription (source) database with the data to be replicated and the replicator engine or Reptor to receive and edit the updates for the destination (target) database are necessary.
Both datahases the subscription and the Reptor contain one replication buffer (RPL), defined with the parameter LRPL. Other buffers, also known from regular databases, play an important role during the replication process, such as the Adabas buffer pool (LBP) and the 1/0 pool (LFIOP) for asynchronous buffer flu shes.
I Subscription
RPL Suffer
l
lBP Buffer!.
-'~ r
110
Buffer
I /J
Replication Engine
-
Oatabase Adabas
=
Raplor Adabas
S V C
-
ASSOfOATA
Figure 5:
,
Mainframe
RPl Butrer D
,
~BPB"ffejj
•,
m
•
110 Buffer
~ RPL
SYSTEM
I
"\S: SLOG
General Requirements - BuHer Usage
In addition to the mandatory Adabas system file, the following files are part of the Replicator Engine (Reptor): • Replication system file that contains the replication parameters, such as descriptions of destinations, subscriptions and the global formats (GFB) . • Subscription log or SLOC file as the overflow area of the replication buffer should also be defined. D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_2, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
2 Different Ways 10 Replicate
14
• A transaction log (TLOG) might also be used for performance and tuning. It is the command log of the Reptor and should not be activated at a regular replicaHan process because of the overhead in the system.
Chapter 3 describes the parameters for each database and chapter 6 explains problems and solutions for incorrect defined buffers and parameters.
2.2
From Adabas 10 Adabas on Ihe Mainframe
The repUcation from an Adabas database on the mainframe 10 another Adabas database on the mainframe is mostly used from companies 10 build a back-up database in real-time for disaster and recovery (DIR). In contrast 10 a mirrored database, the replicated database can be used at any
time in parallel 10 the original database in read-only mode. In a case of DIR all applications can be switched directly to the replicated database and used in read and update mode. The following components are necessary to replicate:
I SubscriptiOf1
Database Adabas
Figure 6:
I
Mainframe
=
Replicator Engloe Adabas
I
Entire
Net·
W""
I :
Target Dalabase Adabas
Replication from Adabas to Adabas on the Mainframe
Source or subscription database
Additional Adabas parameters must be defined and existing ones updated. For more information see chapter 3. Replicator Engine or Reptor
This is a basic Adabas nudeus loading at start·up some additional replication modules. Adabas parameters should be set·up at least with the same values as the source or subscription database. For more information see chapter 3. In addition to the known Adabas system file, the following files are part of the
database • Replication system file that contains descriptions of destinations, subscriptions and the global formats (GFB) .
2.3 From Adabas on the Mainframe 10 Adabas on Linux, UNIX or Windows (LUW)
15
• Optional subscription log or SLOG file as the overflow area of the replication buffer. Entire Net-Work or WCP
Entire Net-Work is also called WCP to differentiate between Net-Work and Network. wep is acting as a broker and passes Adabas direct calls including its buffers to the target database, such as control block, record buffer, and format buffer. For a description see chapter 3 and Figure 12. You must use at this time Entire Net-Work as a broker function to replicate Adabas to Adabas. You cannot use webMethods EntireX. For me it makes more sense to use one broker functio n to replicate Adabas to Adabas and Adabas to nonAdabas databases. In contrast to Entire Net-Work, webMethods EntireX has its own reliable safety mechanism (persistent store). If a connection to the target is no longer available and using Entire Net-Work the transaction will be stored in the subscription log (SLOG) of the Replicator Engine. EntireX will save the transaction in its persistence store, allowing the system to recover to a known point without loss of data.
I had discussions with Software AG and they promised to discuss intemally the approach using also EntireX for Adabas to Adabas replication. Target or destination database
The Adabas target database receives Adabas direct calls from the Replica tor Engine through Entire Net-Work (WCP). Values of parameters and buffers should be set in the same way as for source or subscription database.
2.3
From Adabas on Ihe Mainframe 10 Adabas on Linux, UNIX or Windows (LUW)
The replication from an Adabas database on the mainframe to another Adabas database on Linux, UNIX or Windows is used by companies who use the replicated database to save time and costs for their batch processes. Accessing the replicated database in read-only mode from the Web can also eliminate using networks with all its components between the mainframe and Web servers.
16
2 Different Ways 10 Replicate
The foll owing components are necessary 10 replicate:
I
II
Mainframe
SubScription Database Adabas
Figure 7:
I :
Replicator
Engine Adabas
=
Entire Net·
Wo,
WCP
Network
Firawall ROut8rs Circuil
eie.
II
Linux , UNIX, Windows
Entire Net·
Wo,
WCPIWCL
;=
I
Target Dalabase Adabas
Replication from Adabas 10 Adabas on Linux, UNIX or Windows (LUW)
Source or subscription database
Additional Adabas parameters must be defined and existing ones updated. For mOTe information see chapter 3. Replicator Engine or Reptor
This is a basic Adabas nucleus loading al start·up same additional replication modules. Adabas parameters must be set-up al least with the same values as the source or subscription database. For more information see chapter 3. In addition 10 the known Adabas system file, the following files are part of the
database • Replication system file that contains descriptions of destinations, subscriptions and the global formats (GFB) . • Optional subscription log or 5LOG file as the overflow area of the replication buffer. Entire Net·Work or WCP on the mainframe
Entire Net-Work is also called WCP to differentiate between Net-Work and Network. WCP is acting as a broker and transmit Adabas direct calls induding its buffers from a dient to the remote Adabas target database. The buffers indude the Adabas control block, record buffer, and format buffer. For adescription see chapter 3 and Figure 12. Network
The network with its components, such as lines or circuits, firewalls, routers and load baJancing system, connects WCP on the mainframe with WCP on LUW. Entire Net-Work or WCP on LUW
WCP on LUW ships Adabas calls from a dient to a database through WCL
2.4 From Adabas on the Mainframe to Non-Adabas Databases
17
Entire Net-Work Client or WCL on LUW
An Entire Net-Work Client uses the Entire Net-Work 7 e-business message protocol to access Adabas d atabases. The dient must be installed on any machine from which you wish to access Adabas databases. A Directory Server is needed to store management data for Entire Net-Work in the form of Universal Resource Locator (URL). Target or destination database
The Adabas target database receives Adabas direct calls from the Replicator Engine through Entire Net-Work (Wep). Values of Adabas parameters and buffers should be set in the same way as for the source or subscription d atabase,
2.4
From Adabas on Ihe Mainframe 10 Non·Adabas Dalabases
The replication from an Adabas d atabase on the mainframe to non-Adabas database on Linux, UNIX or Windows is used w ith the replicated database for off-theshelf software or to replicate further with non-Adabas database replication software, The following components are necessary to replica te:
I
II
Mainframe
Subscripton Dalabase Adabas
Figure 8:
=
Replicator Eng ine Adabas
=
"
Entire X Broker
MQSeries Queues
1=
Networl< Firewalls Roulers Circuit ~, .
FirewaHs Routers Cireuit eie.
II 1=
linux, UNIX, Windows E~'
ReplicatOf Target Adapter
I
Targel Dalabase
~ Oracie , SOL, SYBASE, ete,
JL 1=
MQSeries Queues
'GM
Read Queues Save 10 OB
Replication from Adabas to non-Adabas Oatabases
Source or subscription database
Additional Adabas parameters must be defined and existing ones updated. For more information see chapter 3.
18
2 Different Ways 10 Replicate
Replicator Engine cr Reptor
This is a basic Adabas nucleus loading at start-up same additional replication
modules. Adabas parameters must be set-up at least with the same values as the source or subscription database. For more information see chapter 3. In addition 10 the known Adabas system file, the following files are part of the database
• Replication system fil e that contains descriptions of destinations, subscriptions and the global formats (GFB) . • Optional subscription log or SLOG file as the overflow area of the replication buffer.
EntireX Broker (wM EntireX) on the mainframe
EntireX is a message-oriented middleware with broker function. WebSphere MQ (message queue) on the mainframe
MQ is a message-oriented middleware from mM transferring data under the control of a queue manager from point A to point B. One company replicates approximately 50 files from the mainframe over to MQ queues where back-end distributed programs drain the queues and populate to Sybase tables. WebSphere MQ (with IBM WebSphere MQ Client Attachment Feature enabled) can also use the Event Replicator Target Adapter to replicate data to a relational database. Event ReplicatorTarget Adapter on Linux, UNIX, or Windows (LUW) It is necessary to communicate with EntireX Broker on the mainframe and replicate to the non-Adabas target database. The adapter transforms and applies replicated Adabas data to a relational database, such as DB2, MySQL Server, Sybase, or Teradata.
Target or destination database(s)
Any relational da tabase, such as DB2, MySQL Server, Sybase, Teradata, and O rade. Program 10 empty the MQ queues It needs a program to empty the MQ queues and write it to the target database if
WebSphere MQ was only used as a message-oriented middleware (MOM).
3
Definitions and Parameters
3.1
Getting Started
It is necessary to get the following information before YOUf start:
Source database ADARUN parameters, especially • • • • • • •
LBP, the Adabas buffer pool LFIOP, l, parallel LFIOP 1/05 REPLICA TION=YES
Replicator Eogine or Reptor ADARUN parameters, espedally • • • • • • •
LBP, the Adabas buffer pool, at least the same as the source database LFIOP, the Adahas 1/0 pool, l, parallel LFIOP I/Os REPLICA TION==NO
Replicator Engine or Reptor SYSRPTR parameters, especially •
LOGINPUTIRANSACf=70, threshold of RPL buffer before using SLOG
Entire Net-Work (WCP) server ADARUN parameters (mainframe) • PROG=NETWRK • TARGETID= unique within the NET-WORK • NAß= not lower than Reptor OB • LU=250000, not lower • NC= number of commands, for example 2000 Entire Net-work (WCP) dient parameters (Windows) Hardware for Adabas on Linux, UN IX, Windows • •
Server, not a PC Disk arrays I storage area network (SAN), not the C-drive
D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_3, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
20
3 Definitions and Parameters
Bandwidth of the line/circuit between the mainframe and Windows • Depends of the number of bytes and commands to be sent to the target database
• Best should be a ] 00 Mbps duplex circuit. Tests with two 60/70 Mbps circuits using IPSec VPN tunnel were not very successful regarding the speed. To determine the bandwidth of the circuit, the number of bytes passed through Entire Net-Work to the target must be know n. The following Apas/lnsight re-
quest RBLSUM is very helpful to calculate.
-
DBID 000311 RBO RBLSUM
m
,,
..
2010-08-26 TDIZ 13:.f9: 07 Ta 16:"3 111 LlNE
,---------- ---------------- ---------
"" " ,."" """ "" "" """ "" ""
~
,
~
sO,01l5 8,028
357 ,788
77 ,915 36,527 89,335 228,078 353,108
'"
36,770 3,351, 57 0
'"
4,939 7,595 252,291 25 , 243
'"
6 , 314 71 0,645 9,082
.
,~
,
3,515 2 , 444 , 411
~
~
,~
~
~
~
1 0725 2 00 1 1247
8 ,373,20 4 691,U3,928 17, 4 H,475 19,468,891 4,645,420 99 ,213 , 930 43,079,176 45 ,212 64,273,960 269,795,179 29,UO 938,410 1 , 982 , 295 90,377,244 5,149,572 38,402 774,982 H7,611,060 1,271,480
"
~
m
, ----------------, ---------
0.00
86,100,300 10,725 . 00 715.933,188 2,001 .00 33,958,710 435.84 28,381,479 "7 .00 8,933,500 100 . 00 111,530,142 489 . 00 69,lU,9H 193.00 256 ,086 1 , 320.03 1 00,345,330 2,729 .00 991,290,269 2 92.7 9 86,327 1 73.00 7,161,550 1,450.00 2,035,460 268.00 143,011,305 566 . 85 5,477,731 217 .00 64,669 122 . '9 1 ,059,236 1 67.76 1 5 2,6U, 509 214.97 10,135,512 1,116.00 270 . 00 735,950 209.37 2,001,972,609 919.00
, O.
",
'" '"
'"
1524 2729
'" '" '" '" '" ,".
1450 1075
..
1116
'"
'" '" '"
'"
533,955 1 ,7",753,865
0.00 1,OU.DO 1 ,906.00 223.89 533.00 52.00 435.00 122.00 233.05 1 ,748.00 80 .50 60.00
'" "" '" '" "" ".
72.73 122.74 235. 86
'" "
'" "" '" '" "
~
m
,
1043 1906
'" '" " '" '" '" '" " '" '" '"
1749
". ". '" '" '" '" '"
1021
Determine Length of RBL and FBL
Figure 9:
The fo llowing is an example of batch traffic by determining the number and type of commands using the performance monitor Apas/Insight from CA. ~.
~n
100315 100315 100315 100315 100315 10031 5 100315
M
19 19 20 20 20 20 21
Figure 10:
=
"'" ---------- -----------,~
M
~
7B 79 80 81 82
874,107 1,488,111 1 ,517, 170 1, 581,223 1 ,467,842 740,750 11,484
.."
~
~
,-,
,~
~~ ~,
~
" -----------,-" ,• """ ,, ," ,
11.4 19.4 19 .8 20 . 6 19 .1
11.2 18 .9
0.>
0.>
...
Determine Number of Commands
,~
~
~
" --------------755,656 1 ,278, 291 1,324,734 1,396,234 1,318,48a 697 ,74 9 9,958
21
3.1 Getting Started
Update - "Al" D Figure 12:
(ne. d s t l •• s t s uch bandwidth)
Computation for Mbps Bandwidth
Source database in general • How many gigabytes of data (ASSO and DATA) • How many files to be replicated Expeeted traffie • 1,500 update eommands per second are eonsidered to be high • Traffie from online and/or batch Figure 13 shows the used software versions for tests. The target Adabas database was upgraded to V6.2. It also shows the path from .NET Web programs on Windows to aeeess Adabas data on the mainframe using EntireX Broker and RPC Server. lf the replicated database ean be used for read-only Web aecess then EntireX Broker on Windows would be suffident.
22
3 Definitions and Parameters
OB
8.1 .3
6.1.8.03
RPL Buffe,
CSlZOOl
NATN ....... 2.3
Entll'9X RPC Server
,
EntlreX Broker ~~~ 8.0.1
FNAT 4.2.4.3 RPC 6.2.3.1
Wi .... dO\NS
Mainrrarne
Figure 13:
3.2
Used Software Versions
Source or Subscription Database (V8.1.3) Meinffame
Subsctiption Catabase Adabas
RPL Buffer11 LBP BufferU
110
Buffer
I.iJ.
Repliclllioo Engine
...
Reptor Aclabas
s
l~
,,-_
n
RPL BUffer
,
:
~
nLBP BUfferU.
i h;o---l -B~ffer I
\'c7! ASSO/OATAj
Figura 14:
Buffer Usage in Subscription Database
RPl SYSTEM
SLOG
3.2 Sou rce or Subscription Database (V8.1.3)
3.2.1
23
Workload Manager (WLM)
The subscription or source database must have the same priority as the replicator engine (Reptor) . Source SAG: "If an Adabas nudeus gets generally CPU constrained, that is it approaches the
processor capacity limit and if you are not yet ready to try Parallel Services then put this nudeus into SYSSTC or into a low service dass of high importance. Adabas can ill afford to get preempted out of the processor. lf Adabas processes a substantial share of calls originating from batch during high online periods put Adabas into a higher service dass of somewhat lower importance than the response time goals of the TP monitor. However this is not recommended, when the TP monitors concurrently executes transactions, which access other databases like OB2. A high CPU consumption in OB2 could then delay transactions doing Adabas calls! With not much batch activity during online periods Adabas and crcs could be in the same importance service dass."
3.2.2
ADARUN Parameters
Besides the most important parameters for performance (LBP, LFIOP, NAB, ASYTVS and FMXIO), the following parameters must be defined in addition for the replication process: Source database ADARUN parameters, especially •
ADARUN REPLICA TION=YES
This parameter loads at start time of the Adabas nudeus the repIication moduleAOARPN . •
ADARUN LRPL=700M.
The RPL buHer holds the compressed data to be replicated and should be defined larger than LBP and large enough to avoid an overflow because on the source database does not exist an overflow area (SLOG). • RPLSORT=YES (default) Transaction data are sorted by a key consisting of the fUe number, ISN, and relative update number. All modifications for a given ISN in a file are consolidated before replication occurs. If you do not want to sort to occur, specify RPLSORT=NO.
• RPWARNINCREMENT=1-99 (10 default) Warn increment that indicates that the replication pool usage has exceeded the threshold set by the RPWARNPERCENT parameter. • RPWARNINTERVAL=xxx (60 seconds default) Suppress waming messages after xxx seconds.
24 •
3 Definitions and Parameters RPWARNMESSAGELfMIT=xxx (5 default)
The number of RPL pool usage waming messages that can be issued to the console before message suppression begins. •
RPWARNPERCENT=O-99 (0 default) The threshold for RPL pool usage at which warning messages should be sent.
Appendix, chapter 9.1 and 9.2 shows the ADARUN parameters of a source database used by a company with 57 subscriptions spread across 4 LPARs running on 8 separa te Replication STes.
3.2.3
ADADBS Replicate Files
Run also the ADADBS utility to turn on the files to be replicated and turn off the overhead of the before image log (DSBI). ADADBS REPLICATION FILE=006,ON,T ARGET~302,DSBI=OFF ADADBS REPLICA TION FILE=fnr,ON,T ARGET=reptor-db,DSB1=OFF ADADBS REPLICA TION FILE=fnr,ON,T ARGET=reptor-db,DSBI=OFF ete.
The Adabas report displays the repUcation status of the files and how the before image of updates has been set. In the below example, DSBI was set to the default value ON: 'Collect before images of updates = Yes: In later tests, it was set to OFF. e.nip) .................................
• • • ••••••••••••••••••••••••••••••••• ,
(ST-ADA-CALC-EIlAT)
2010-03-0~
13:5~ : 15
LOB f i l . No Cont-in LOB fi. ld . NO RPLUPDATEONLI
No
Fil. i. replicated : Raplicator targ. t
Coll. ct hafo.,. e.nip)
Figure 15:
ID
~9..
ot update.
• I ••
302
Replication Status in Adabas Report
It is very time consuming to manually check the report of the replication settings
for example 200 files. I wrote a parse program in Natural, which displays these flags in areport. See also the chapter describing self-written monitor tools: 7.2.10.
3.3 Replica lor Engine - Replor (V8.1.3N3.2.1)
3.3
25
Replicator Engine - Reptor (V8.1.3N3.2.1) Mainframe
SubSCfiplion Database Adabas RPl Buttern LBP BulferU
r--
=r~
~ /1
Buffer
Figure 16:
3.3.1
-
RepliCation Engine Replor Adabas RPl Buffer
,-
,
•, •
m
n
~LBPBi~U '0
Buffet
.'%-LL
\\
ASSO / DATA
SYSTEM
RPL
'\\ SLOG
Buffer Usage in Reptor
Workload Manager (WLM)
The source database must have the same priority as the replicator engine (Reptor). SourceSAG: "lf an Adabas nudeus gets generally CPU constrained, that is it approaches the processor capacity limit and if you are not yet ready to try Parallel Services then put this nucleus into SYSSTC or into a low service dass of high importance. Adabas can ill afford to get preempted out of the processor. lf Adabas processes a substantial share of calls originating from Batch during high online periods put Adabas into a higher service dass of somewhat lower importance than the response time goals of the TP monitor. However this is not recommended, when the TP monitors concurrently executes transactions, which access other databases like OB2. A high CPU consumption in OB2 could then delay transactions doing Adabas calls! With not much batch activity during online periods Adabas and CICS could be in the same importance service dass."
26
3 Definitions and Parameters
3.3.2
SLOG File
The subscription log file (SLOC) is an Adabas file and is used as an overflow area if the threshold setting of Reptor's RPL buffer (LRPL) is reached. The size should be targe enough to hold a couple of days not replicated updates (transactions) if the destination is not available. Many companies reserve an entire 3390 pack for theSLOG. 2011-01-05
. . . . . A D A B A S BASIC SERVICBS . . . . . - Di . pIey ril" wyout -
13 :59: 16 DBtD 312
**** •• • •• •• •• •• •
PDIU'042
•••••••••••••••• Reoord
i12665
TOP ISM Ha>< ISN . _ " , ud Minimwo ISM Si". or ISN ..•.. NWIIbe" of Updat.. ISN Reu •• q a .
912661 20000079
Sp • .,.
R100P; 3/B/Ol
LPIOP-13000000 L1:_ 12000
LEN 01'
~w
~
ALLOW ACCESS !'ROM OTHBR NBT-WOP./t NODBS
~w
LOGGI_nS
LOG BASIC ADALOO RECORO
~w
~m
~w
""_NO
~w ~w
~w
~w
urne"
POOL
BUI'R FLUSH POOL
LEN :IBN LJ:ST 'fABLE
JICB
LOGm-ns
LOGSB-YES
~w
~m
~w
LOGla-NO
~w ~w
LOGIo-OO LP-8000
•
~w
LWlU'2_S000
WORJI> PART 2 JD 1/1B/l
~w
LQ0020000
L1!NGTH 01' SEQUEN'l'IAL
SORT
~w
L5_150000 L0-200000
~w
1MP-500000
LENGTH 01' WOlUI: POOL
~w
+
ADARUN M:lDE-e«JLTI
PRQ'rECTJ:ON BLOC:KS -
~
40>60>150
WOJUI;:
PART 1 - .;ro 1/18/1
~
~R
TJIBLZ
RPULD 2/17/10
INTBRMlWIATII USER Bt1FFER MULTI roll. V62
ADARUN NABa420
• ATTACHED BUTTERS 30>50 NSH CLASS 07/02/96
ADARUN NCW120
NUMIU!R 01' CQZ' 5
ADARUN NH_20000
NUKBER 01' HQE' S ALL USBRS
ADARUN NJ:$NIIQ-SOOO
•
ADARUN NONOES-NO
ADARUN N'I'W15
OON-OESClUPTOR S&ARCIIl!:S 03/03/03 ~ OF ACTrvz CIO'S PER USER ~ OF THREAOS
ADARUN NU>o4000
NUKBER OF USERS
ADARUN QPENRQ-NO
---> --->
ADARUN HQCID-SO
ADARUN PLOGRQ-NO
HQE'S ZACH USER
DEFAULT DEFAULT. TBS
~w~
ADARUN QElLKSrz_ 32760 ADARUN SVC-241 ADARUN =-900
ADARUN TLSCMD_1800 ADARUN
~lBOO
TRANS TIME LDUT T:IME LDUT I'OR $X cotea.NDS OON- ACTIVITY T:IME LIMIT
ADARUN THAE_ 1800
OON-ACTIVJ:TY T:IME LJ:MIT
ADARUN UEX,200EJ:IT2R
HANDLJ: BOTH P AN!) C LOGS
ADARUN
O~ _ 3390,DUALCLS _ 15000
"
BS-PRINTBR-PROPERTIES
,
-
EVENT REPLlCATOR SUBSYSTEM . . . . .
Liat of Predict ril ••
IFDIC-(lOO,201)
BS-P~NTER-PROPERTIES-ADA
BS-ST-RBQ-LOG BS-ST~-REQUEST
DIETBR-USER-REQUEST CIS-CNTL-PROCESS CIS-RUN-CONTP;OL CIS-RUN-OONTROL-AUDIT etI-REASON-CODE etI-REASON-CODE-LOG etI-REVIZIi'-GROUP etI-RBVIZW-GROUP-LOG
U A A A A U U U U U U U
2009 - 01 - 09 M-PJ:CIUl'J:
Maat.r 1'11.
--------------------------------
BS-PRINTBR-PROPERTIES-ADA
CP-CON'l'ROL-ADA CP-RUN-ADA CP-IWN-LOG-ADA ST-REVIZIi'-ADA ST-RZVIZW-LOG-ADA ST-REVIZIi'-ADA ST-RZVIZW-LOG-ADA
COIOIIIand _>
Ent.r - PP1 --- P!'2 --- PP3 ---PP4 --- PP5--- PP6 --- PP1 --- PP8 ---PP9 --- PP10 --PPll -- PP12 --r.xi t
Figure 22:
Accpt
+
List and Select Predict Files Using Replicator Subsystem
Moonu
3.3 Replica lo r Engine - Replor (V8.1.3N3.2.1)
33
The next screen shows the Predict Parameter. 10,4& , 29 ..... A 0 A B A S rDIC- (100,201)
----------
• DIETER-VSER-REQUEST
•
Gen.ration
...
2009 - ()1 - 09 M-RP1121
(0:100590
IOn N _ •..•.. l'il.ID ••••••• T.r'l.t ~il. ID
EVENT REPLICATOR SUBS!ST!!M . . . . .
In~ormation
Va.r
,
D.~
_
..
DBIO
~"
----------
Adab . . . . . ."'. ion Oc:curr.n..... u • .cI ,..,11 ~ormat . .
c-nd _ >
Enbar-PP1 --- PP2 ---PP] --- Pr4 --- PP5 ---PP 6--- PP7 --- PP8 --- PP9--- PPI0 -- PFlI -- PP12 --H.1p Exit zx.c S.l Unl"" Man\!
Figure 23:
Select a Predict Files Using Replicalor Subsystem
Press the -key to EXEC the generation of the GFB for this file number. 10,50 : 48 ..... A 0 A B A S I'DIC- (100 ,201 ) 10100590
Gd N _ . .
Pil. ID .. . T.r'l.t ~il. ID
----------
EVENT REPLICATOR SUBS!ST!!M . . . . .
• ST-AOA-CALC-EOA'l'Z
•
Gen.ration
In~ormation
+-----------------------------+ Maba. V.raion ...• 17 I Oc:cur",. nc •• u • .cI . . • ! I TU11 ~ormat ..•...• !
( ! /N)
+-----------------------------+
----------
_
..
Adab . . .... r .ion Oc:curr .n..... u • .cI .. Pull format: . .
Figure 24:
Select a Predict File Using Replicator Subsystem
Accept everything and press the <ENTER>-key.
2001l-01-09 M-RP1121
3 Definitions and Parameters
34
ARrO0132 ,
ro~t
1 0:52:37
buLLe "
~ . n. r.~d
..... A D A B A S
. u~ c.
•• Eully
EVBNT RBPLlCATOP. SUBsrsTZM . . . . .
Global
r o~t
•
anNam.
.,
,Co
L Fi. ld ID
,,, ,, LAS'r - CHGD-DATZ-TlME ,
, • •• •• •• •
LOG-COUNTZR BN'l'J:TY
LAST-eRGO-USER
CALC - CODI!:
SlIT-DAß
1 En-DAß-9C 1
JU:P-DATJ:
1 DATA
C-nd
BuEt. "
~n9th
..
15 . 0 50 . 0 15 . 0
' .0
10 . 0
' .0 ' .0 ' .0
250 . 0
2009-01-09 N-RP1122
1 of: 9 Fon>a t BuE!'. "
M,B,i" Q,50,A AC,15,A AD,8,A JJ!:,lO,A
Ar,B,A AG,B,A AB,S,A AI,250,A
- >
Bnte,,-PFI---PF2---PF3---PFf---PF5---PF6---PF7---PF8---PF9---PF10--PF11--PF12---
•
He lp
Figure 25:
Save the 5elected Predict File Using Replicator Subsystem
After the format buffer was generated successfully (see ARF00132 message at the top), press the -key to save. The message' ARFOO022: New GFB added' will be displayed. Press the -key and the new GFB has been added to the list of global format buHers. 10:55
,.,
"
.h._ A
tI
A 6 A S
EIlEN'!' U5'Ll CA'1'OR SUßST8TI!:M H •••
N_
------------
S. 1
N.....
8.1
N. . .
2009-01-09 N-RI>1l30
Li.t ot G1ob.1 Pormat Butt.r.
8. 1
N .....
G100589 G100590 G100591 G100592 G100593 G10059
Ent. r-5'P1---5'P2---5'P3---5'P0,
CoaImand Queue Hold Queue C1ient Quaue HQ Uaer Liait Threada Workpoo1 15M Sort Compl_ Search Attached Burrer
.-
In Uae
on 12-,J7,N-2011 20:29:09
67,490 2,556
,, ,
.",
Total To Pree $pace Wri te Liait Moditied
Adabas Windows Buffer Pool Statistics
5\ ) : 0\ ) :
20,971,500 665,600
.""
48
3 Definitions and Parameters
3.9
EntireX Broker (V8.0.1)
webMethods EntireX Broker is needed replicating Adabas data from the mainframe to non-Adabas data on Linux, UNIX and Windows (LUW). The Event Replicator Target Adapter is also necessary on LUW to communicate with the EntireX Broker. Depends on the target database, different target adapters must be used. Figure 46 shows the target adapter for SQL databases. EntireX Broker is also used to access Adabas data located on the mainframe with a program located On Windows, for example .NET. On the mainframe site is also a Natural RPC server necessary. But this is not necessary for the replication process. Please see a parameter example of EntireX Broker V8.0.1 in chapter 8.9, EntireX V8.0.1 Parameters. See graphie for RPC proeess Figure 12.
3.9.1
EntireX Broker (V7.3.4)
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• • Entir.x Broker Attribute Pile .,' .3 • •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
........................................................................ 'MOOULII : ETBAHHZ 'SOURCE !.OCATION: HK. ENTIR& . V134 .AT'l'RB . LOCSRCE ' ruNCTION:
lIND(:
(ETBAHHZ )
.----------------------------------------------------------------------. •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• • Broke r
ape • HOST _ • KBY-LABBL _ • TRUST-STORE - <tru. t_. tora> • RESTART _ YBSlno • RETRY-LIMIT _ UNLDtln • RETRY-TDIZ _ lOS • REUSE-lUIDRESS y •• INO • VBRIFI-CLIBNT _ Y•• INO
. . . . . . . . . . . . . . ,,7.3: TCP p.r..... tar . DEI"JU1LTS _ TCP HOST _ h .. dc.l.c.h ..... t . 9" PORT _ 50513 STACK- NANI!! _ HWTCP RESTART • IES RETRY-LIMIT • 20 RETRI-TDIZ _ 3 CONNECTION-NONACT IH REUSE-lUIDRESS - YES
.......... . .......... . ....... . .... .
50
3 Definitions and Parameters
DEI!'AULTS • SECURlTY . . . . . . . . . . . . . . . 7.3 : S.rvic. d.finition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DBI!'AULTS • SBRVICE CONV-LDnT CONV-NONACT LONG-BurRR-LDnT MJIX-UOWS NOTlrr-EOC SERVER-NONACT SHQRT-BurnR-LDnT TPANSLATlON UOW-MSGS
• RRl" 07/20/05 Ch.n'J..:i t'rOlll. 248
• . n . ur. EOC not.ific.tion • RRl" 07/20/05 ch.n'J..:i t'rom 2008 ~~
• CNH 05/09/00 dem' t
SAGTCHA
r.quir. tr ... d.tion
'00
MII t SVRSC, MIItSVRSC , MIItSVRSC, MIIZSVRSC,
SBRVICE SBRVICE SBRVICE SBRVICE
._--------------------------------------------------------------------_. Rap1ic.t.or S.rvic. Definit.ion •
._--------------------------------------------------------------------_. u ..... u
•• u • •7.3
: S.rvic. definition. u •• u . . . . . . . . . u •• u . . . . . . . . .
DBI!'AULTS • SBRVICE CONV-LDnT CONV-NONACT LONG-BUFFER-LDnT MJIX-UOWS NOTlrr-EOC SERVER-NONACT SHORT-BurnR-LDnT TPANSLATlON DEFERRED UOW-MSGS SERVER SBRVER
~~
...". ...'"
1208
~,.
50000
• bump up per Victor 6/12/09
•
Al . .. - .... tch broker ...... 1 • . n . ur. EOC not.ific.tion • lU. ..
~,.
~
REPLlCATION R1!PLlCATlON
SERVICE SBRVlCE
AA'
INlT
._--------------------------------------------------------------------_. Short.ar Con ... Non-act.i ... ity t.'--out. . . .d MIlZ Tr.na1.tion Routin • . . . . . . . . U . . . . . 7.3 : S.rvic. definition . . . . . . . . . . . . . . . . . u . . . . . . . . . . . . DBFJWLTS • SBRVICE CONV-LDnT CONV-NONACT LONG-BurRR-LDnT MJIX-UOWS NOTlrr-EOC SERVER-NONACT SHQRT-BurnR-LDnT TPANSLATION UOW-MSGS
~~
••
~,.
>0, HO 2008 ~~
~~
'"
• MIIt Tran . 1ation Routina
MIIZSVRTD, SERVICE MIIZSVRTD, SBRVICE MIItSVRTU , SBRVICE MII t SVRTD, SBRVICE
~~
PING
3.9 EntireX Broker (V8.0.1 )
51
.
.---------------------------------------------------------------------_ ---------------------------------------------------------------------_. • ._--------------------------------------------------------------------_. • ._--------------------------------------------------------------------_.
. •
.
•••••••••••• • • v7.3: Servie. deEinition a •••••••••••••••••••••••••••••• DE FAULTS _ SERVICE CCINV-LIMIT CCINV-NQNACT LONQ-BUFrER-LIKIT NOTIrr-EOC SERVER-NQNACT SHOlI.T-BUITER-LINIT TRANSLATION
•
~,
•• m ,.
~,.
~,. ,~
ACLASS, SERVZR BeLASS, SERVZR CCLASS, SERVZR
ASERVER, SERVICl! BSERVER, SERVICII: CSERVER, SERVICl!
ASERVICl! BSERVICII: CSERVICl!
._--------------------------------------------------------------------_. • Servie. apeciEic Atuibutea / Tutorial Non-Con ... araational Sarvice • _--------------------------------------------------------------------_. •
.
•••••••••••••• v7 . 3: Servie. deEinition • •••••••••••••••••••••••••••••• DEFAULTS _ SERVJ:CE CCINV-NONACT SERVER-NONACT TRANSLATION
,. ,. SAGTCHA
CLASS-ETB,SERVER-TUtori.l,SERVICE-N~ltiSrvl CLASS _ETB,SE~tori.l,SERVlCE-N~ltiSrv2
•
CLASS-ETB,SERVER-TUtori.l,SERVICE-NcHoaeply CLASS_ ETB, SE~tori . l , SERVlCE-Nclfi thRep11' CLASS-ETB, SERVER-Tutori.l , SERVICE-NcHoaeplyAal'
.----------------------------------------------------------------------. • .----------------------------------------------------------------------. •
•••••••••••••• v7.3: Sarvie. deEinition a •••••••••••••••••••••••••••••• DEFAULTS _ SERVICE CCINV-NONACT SERVER-NONACT TRANSLATION' CLASS- ETB,SERVER-TUtorial, SERVICJ:-CvS-ckVround CLASS-ETB, SERVER-TUtori.l , SERVICZ-CVOa tePromSrvAa y CLASS- ETB,SERVER-TUtoria l,SERVICJ:-CvDa taToSrvAal' CLASS-ETB, SERVER-TUtori.l , SERVIcz-cvo.mooeravOuiea CLASS_ ETB, SE~tori.l, SERVI~Hold CLASS-ETB,SERVER-TUtori.l,SERVICJ:-CvPar.lleLAay
•
CLASS _ETB,SE~tori . l,SERVlCE-C"UnitOfWork
• _____________________________________________________ - - - - - - - - - - - ______ 0
•
0 ______ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ _ _______________ •
•
. . . . . . . . . . . . .....7.3: Service deEinition a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
52
3 Definitions and Parameters
DB~TS
• SERVICB
CONV- NONACT NOTln-BOC SERVZR- NONACT TRANSLATION
8M
CLASSooE'l'B , SBRVER-Tu tori.l ,SERVlCEooRaqu•• t CLASS _B'l'B,SERVER-Tgtori.l,SERVI~.it
*----------------------------------------------------------------------* • 5.rvi.,. Attribut.a / NATUPAL S.rvic•• *----------------------------------------------------------------------* .~"if"i"
P.PC
.. . .. . . . . . . .. . ,,7.3: Service d.tinitiona .. . . .. . . . . . .. . . .. . . .. . .. .... . . DIU'lWLTS _ SBRVICB COHV-NONAC'r 5M NOTln-BOC NO SERVZR-NONACT TPANSLATION CLASSooRPC , CLASSooRPC ,
SBRVBR-SRVl ,
SUVlCE-CALLNAT
SBRVBR-MHZSRVl,
SBRVlCE-C~ A T
*• ._--------------------------------------------------------------------* ZN'I'IRE/x Attach Manaqer t..t •• rvi" •• *._--------------------------------------------------------------------* .. . .. . . . ..... .
v7 . 3 : S.r'I'ic. definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
_ SBRVICB CONV-NONACT 5M NOTIrY-BOC NO SERWR-NONACT TRANSLATION OAOTCKA
DB~TS
,.
-,-, ,_. ,_.
''''''' ''''''' ''''''' '''''''
ATCLASS. SERVER ~,
SERVER
~,
'''''''
SERVER
'-, ATTBST,
SERVICE SERVICE ATTBST, SERVICE ATTBST, -,~ ATTBST, SERVICE
ATTACH
CBCSRCH SDLDSTC SDLDINTl SDLDINT2
._--------------------------------------------------------------------_. END OF DEFINITIONS ._-------------------------------------------------------------------- _ •• • •••• •• •••• • •••••• END . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Figure 47:
3,9,2
.
webMethods EntireX Parameters (V7.3.4)
Event Replicator Target Adapter
The Event Replicator Target Adapter transforms and applies replicated data to one relational database at a time, such as D82, MySQL, Oracle, SQL Server, Sybase, or Taradata.
3.9 EntireX Broker (V8.0.1 )
53
Example for SQL Database '.qlr.p 'Mon Feh 07 03:15:20 PST 2011 .ource.r•• rv. r.~/RBPLlCATION/ART del.tebeLor.in •• r t- tru. me •• aqepr. t .tchcount-20 cLo.t.ba ••• ba tchcoun t-O under.cor. . tru • • ource .captur. _ Lal •• aql.rrorcontinu. .true oompatib1. 23-ta1 •• cLo.t.ba ••. nu11tortypedecima1_ t.1a. occurr.nce- tru. webda". c.ch. ai z . .25 . tartup.bui1dn~r-1048
aqla.rv.r.portn~r-l'33
. ou rce.maxr.c. i". 1.n-32768 aqlr.p.".rbo •• _O da t.ba ••. conn. ctionr.u. _tru. aql •• rv. r . ho.tn...-.c.!.haqlc3,,1 a ource.tr.c. .O brok.r. ra.rv.r.ddrw>utxx brok.r . captura_Lal •• a tartup. count-603 quo taa trinqwtru. dat.ba •• . dbn.....-ADT aourca.id-hwdc . 1 .cahwn.t . qov\ 50513 brok. r.tr.c. .O brok. r . maxr. c . i". 1 . n_ 32768 brok.r. idaxxxx . ql •• ~ . r . pa .a worg-xxxx 1ow.rc. . . .t.l •• cLo.t.ba •• . u •• r d*Linaamp tyva 1u._ tru. brok.r . • • ittimaw20S . ql •• ~ . r . u •• rn. . . .ourca.akipconn.ctionwL.l •• • ql •• rv. r . dbn.....-Adaba . cLo.t.baa •. nul l tortypatloat-L.l •• brok.r ••• rv.r.ddr-xxxx aourca . aourcatYP'"'BROI'ZR
_.a
Figure 48:
3.9.3
Parameters for Target Adapter fOT SQL Database
Schema and Replica!ed Da!a
The XML schema for the data is created from the generated field table (GFB) for the subscription that processed the replicated data. Each of the create, insert, update or delete operation will create the SQL table in the relational database management system (RDBMS). Generated Field Table GFFT
The fjeld table is created when you generate a global format buffer (GFB) for a subscription in Event Replicator for Adabas. See chapter 3.3.6. It is generated from Predict file definitions.
54
3 Definitions and Parameters
,•
---
L OB N _
, ,••
--------------
M
top -------------
PE.RSONNBL-ID
,, M _
•• • •
AB ~
PIRST-NAME
AI)
MIDDLZ-I
~
MAR-$TAT
.....
-"
• ---------- , ----------B B B B B
~,m
0
,
0
"• "• • N N
0
S_SINGLB D~~ _~D
, , ,m ,"" '" , , ,• , 1 AG SEX
I'ULL-AI)DRJ:SS
B B B
~
BA AL
B
~,
TELEPHONE AN ARn-CODE 10M PHONE BD ~ ,JOB-TITLE
B B B B
BD LE./WB-DUB BV LEIWB-TlUIEN
N N
~
, ,••• ., ,
,.n
LBAVE-DATA
Figure 49:
•, "", • "• " ' .0
N 0 N N N N
0 N 0
' .0 N
Generate Field Table - Source: SAG
EXAMPLE "Create":
~'10'>
nam._ 'DBPT' . i_ 'lS' f . t - ' .trinq' 1'ro-'O' 1. n _ '6 ' k .y-' ky'/> n . . . . 'JOB- TITLB' . i.'16' fmt-' . tring' pro-'G' l . n_'25' k .Y"' ky'/> n . . . . 'CURR-OODB' .i.'18' fmt-'.trinq ' 1'ro-'O' 1.n. 'J' att-'~'
"", _ _ ' :INCOMI:' />
Figure 51:
Generated XML Schema ' Insert' - Source: SAG
EXAMPLE "Update":
Figure 53:
Generated XML Schema 'Delete' - Sou rce: SAG
4
Compare Files of Source and Target
Despite users' statements that the replication works correctly and comparisons between SQurce and target are not necessary, self-written comparison programs sometimes show differences. Many times Adabas response codes are the reason that records oe ISNs are not replicated. The system log (DDPRINT) of the subscription database and the Replicator Engine shows response codes but normally, there is no routine in place to check for erfOTS. Chapter 6 points out the reasons and chapter 7 describes the monitor routmes to detect response codes, errors and high water marks and how to notify the relevant grOll ps .
-
• '1\_.
Rap l leated Productlon
Produ etion Oatabas.
D at.base
ASSO
ASSO DATA WORK
DATA
W ORK
? Figure 54:
Compare Source and Target Files - Amount Fields
Three methods will be discussed how to compare source and target files to be sure that both are identical.
D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_4, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
58
4.1
4 Compare Files of Source and Target
Compare Amount Fields
Counting amount fields of files in both subscription and target databases and compare the result is easy to accomplish. Appendix, chapter 9.5.8 and Figure 211 show the Natural program. 11.1. RBAD HULTI-FBTCH 'MULTIFBTCH CONTRIB ADD POST-TAlI:-CN'l'RB-AMT TO 'POST-TAlI:-CN'l'RB-AMT ADD l'RE-TAlI:-CNTRB-1oH'l' TO 'l'RE-TAlI:-CNTRB-AMT ADD DB-POST-TAlI:-CNTRB-AMT TO 'DB-POST-TAlI:-CNTRB-AMT ADD DB-PRE-TAX-CNTRB-AMT TO 'DB-PRE-TAX-CNTRB-AMT ADD LOG-COUNTI!R
Figure 55:
Natural Program to Count Amount Fields
WINDOWS :
CONTRIBU'I'IDN FILZ CHECK Smeo.R.Y POST-TAlI:-CN'l'RB-AMT 27 , 673 , 968 59 PRII:-TAX-CNTRB-AMT 29,942,205,909 . 37 DB-POST-TAlI:-CNTRB-1oH'l' 26 , 347 , 293.89 DB-PRII:-TAX-CNTRB-AMT 25,380,190,817 . 46 LOG-~ 839,8B7 , 724 RECORDS REl\D: 146,223,112 MAINriUIME :
CONTRIBU'I'ION FILZ CHBCK SUIoeQI.RY POST-TAlI:-CN'l'RB-AMT 27,658,136 50 pg-TAlI:-CNTRB-1oH'l' 29 , 942 , 205,909 37 D8-P08T-TAX-CNTRB-AMT 26,331,461 80 DB-PRE-TAX-CNTRB-AMT 25 , 380,189,01 •.•• LOG-~ 140,074,151 noad _ith Hu1ti:r. tch Record8 Read : 146 , 223,098
Figure 56:
'"
Result of Natural Program to Count Amount Fields
Figure 56 shows that the file on the target database (Windows) counts 14 records more than the file of the subscription database (mainfrarne). It can be that the file on the target database was not empty before the initial-state or a replay function was restarted.
4.2 AP! USR4Q11N - User Exit
4.2
59
API USR401 1N - User Exil
Many files don't contain amount fields and other comparisons must be found. Fortunately, Natural has the capability with its APIs. The program CHECKSUM uses Natural's ArI USR4011N to compare two Adabas files. 0010 8ampl. CHECKSUM 0020 1 Provid. OBIO. 0030 Pro .. id. f'i1. name in 'r:1L11: provid. Fil. Number in 'FILII:-NBR 0040 Pro .. id. f'i1. name in VIEW 0050 Provid. • tull OOM definition to11owinq VIEW 0060 Pro .. id. th8 LlU:CL in 'OOM 0070 0080 0090 0O", ot CHECK8UM aa CHltSM006 I'ile 6 0100 OBrINB DA1'A LOCAl. /. « « < INU' 0110 'OBI (NS) (N5) /. « « < INI1' 0120 INI,. /. « « < 0130 'I'ILII:- NBR (NS) INI1' /. ««< 0140 'FILII: (Al2) VIEW 81'- NC- CALC- EDATB /. « « < 0150 IoOG-COUNTBR (PI5 .0 ) /. « « < 0160 0170 ENTI1'Y (ASO) LAST-CHGD-OATB-TIME (AlS) 0180 LAST- CHGD - USER 0190 (A08) CALC-CODB (AlO) 0200 E!T-DATE (AOS) 0210 B.... -DA'l'B-9C (A08) 0220 UP-DATE (AOS) 0230 (A250) 0240 Rl!:I)EPINB OOM 0250 0260 'DOM (Al65) ««< 6
,, ,•
•
•
..
,,, '" ,, ,, ,, ,,, ,,
,,, , ,•
,
0490 0500 0510 0520 0530 0540 0550 0560 0570 0580 0590 0600 0610 0620 0630 0640 0650 0660 0670 0680
m.
,
'*
OBI'INB SUBR()UTINB HASH-R'l'N ~1' "USR1040N " '1040 ASSIGN 14011.1'UNC _ 1 ~1' "USR4011N" 14011 ASSYGN 14011.1'UNC 2 ASSIGN 11023.MS *CPU-1'lMB R. RB10D MUL1'I-rBTCH ON OOM 8Y ISN ASSIGN 14011 ."rI!:XT _ 'OOM ~T "USR4011N " '4011 END- RB1oD ASSIGN 14011.1'UNC _ 3 ~1' "USR4011N" 14011 ASS:1GN 11023.MS • •CPU-TIME - 11023 . MS ~1' "USR1023N" 11023 DI8PLAY '08:10' 11040 . 08 'Recorda' *COUNTZR (R.) 'Ch.ck.um' '4011.HASH 'CPU 1'~' '1023.1'rHB (EH-HH :1I:SS.1') BND-SUBROU'l'INB
Figure 57:
Result of Natural Program to Count Amount Fields
Appendix, chapter 9.5.7 shows the entire Natural program 'CHECKSUM'.
60
4 Compare Files of Source and Target
..Database 300
ASSO
DATA
WORK
Figure 58:
Database 400
ASSO
OATA
WORK
Result of Natural Checksum Program
The program reads all records of the source and the target database, creates checksums and number of records. 80th number of records and the checksums must be identical, see Figure 58.
4.3
Compare Highest ISNs (TOPISN)
Anather way to make su re that the source and target files are the same is to compare the number of records of both files. But this won't be sufficient because you can have the same number of records in a file but the highest used ISNs are different; and this leads to erraT messages (RSP 113) during the replication process. Also, to compare the MAXISN won't work. MAXISN on both source and target can be the same but the number of records loaded can be different. The Adabas report shows both the number of loaded records and the maximal used ISNs of each file . Manually comparing Adabas reports of 200 files and more can be too time-consuming . At first, I wrote a parse program to compare programmatically the two reports. I included also in a spread sheet the TOPISN, REUSEISN and MAXISN fields. It showed in later tests that if the target files are not initially loaded with REUSEISN, it is a problem and leads to errors.
61
4.3 Compare Highest ISNs (TOPISN)
Source Database - Adabas on Mamframe File Records Number Loaded
122
12,434, 148
Figure 59:
TOP·ISN 12,440,
973
Target Database - Adabas on Wmdows
MAX·ISN
ISN Records Reusage Loaded
12,501 ,
Ye,
215
12,434, 148
TOP-ISN 12,440,
960
MAX·ISN
ISN Reusage
12,449,
Enabled
279
Compare Highest ISN (TOPISN) and Others
The disadvantage of using a full Adabas report in batch is that it can take up to 30 mmutes to finish the job with 200 fil es. In addition, some organizations must pay for used CPU time and I/Os. Source OB Mainframe 09:58:46 DBID 39
•••••••••••••••• • Fil. 122 ••••••••••••••••
·····ADABAS - Di.p1ay ST-UVIZW-LOG
Rec:orcb lO.&.c:l 12434148 TOP ISN . . . . . . . . . . . . 12440973 Ha .. ISN • ..peete
Figure 60:
Adabas Report Records Loaded, TOPISN and MAXISN (Mainframe)
Target database Windows Databa •• 2S1, Fi1.
122
(ST-REVIEW-LOG
Hi",h.a t Inc:le>< Level: 4 Top ISN: 12,440,960 R.eord. loaded: 12,434,148 La . t FOT ~fication: I8N .,.,uaa Figure 61:
26-~-2011
Paddin9 Faetora:
ASSO
Maxim". ISN • ..peetec:l: 03-~-2011
10 : 24:47
5\ , DATlI.
5\
12,449,279
11 : 21 : 59.055000
$pace r e u . a",e: Ciph.rin",:
Enabl .... Di.abl ....
Adabas Report Records Loaded, TOPISN and MAXISN (Windows)
Ouring my researches, I came along with a Natural program written and published on SAG-L in 1999 by Lewis Pritchard that determines the highest ISN in a file. Since there is neither an equivaIent Adabas nor a Natural function available (get the highest used ISN), I used this program and adapted it for my dient. Now, the Natural program uses the Adabas command 'Ll' with an 'F' in command option 2 to get the next highest, unused ISN (thanks to Larry Frazin).
62
4 Compare Files of Source and Target
The first parameter is the DSIO, the second parameter the startmg file number and the third parameter the number of files to be reported. 11 * -------------------------------------------------------------1/* D.teno.ine the hi'lh •• t u •• d ISNa per ~il. 11 * ISHHICH 112 ,1,10 1/* --- - -11 * I I 1/* DBID
+----------
"+
I I +------
1/* ,,>
start with t i l . nUJaber
+---- nwober ot fil •• to be reported / ' + ------ - ---------- - --------------------- - ---------- - -----------
1/*
IIISNHIGH
IUlEC NA'l'ßT ,OJ:PT-ST ,tI~EVL, SISOUT-+ ,Tnm_lOO, PRM-' IMootI ,MADlOooO • MAXCLaO , _ 0 ,LTaggngg , AtJ'l'OooON, DUaJIBEND' IISy.SOUT PD SISQUTw* t!CMPRTOl OD SISOUToo* IlslSIN PD. LOQON DIETER
If
".
ISNHIGH 112,1.10
"
Figure 62:
~
Compare Highest Used ISN OCL)
ISNIIIGH
Paq_ D1HI):
DBID: DRID: DBID: DBID: DBID: DBID: DBID: DBID: DBID:
Figure 63:
,
m m m m m u> m u> m u>
112 , 1 ,10
LIB-DIBTI!:R
03/28/11 ~,
~ , ~,
~ , ~,
~ , ~,
~ , ~,
~ ,
,, ,• ,,,• "
1 H:IGHJ:ST RIGHJ:ST RIGHJ:ST RIGHJ:ST RIGHJ:ST HIGHJ:ST HIGHJ:ST HIGHKST HIGHJ:ST HIGHBST
ISN: ISN: ISN: ISN: ISN: ISH: ISN: ISH: ISN: ISB:
08:14:2.
'", '" 1145 '" U5690 '"
llO0997
'""
Compare Highest Used ISN (Output)
The Natural program ISNHIGH is located in appendix, chapter 9.5.10.
5
Recovery After a Crash
Normally, the Event Replicator Server restarts automatically after an abnormal end and is able to recover any lost replication. But in same cases, data is lost and the replication must be replayed, as reported in chapter 6 with some response codes from the target database and Entire Net-Work. The new ADARPL utility and its REPLA Y function is similar to the known regenerate function of ADARES. ADARPL needs the protection log of the subscription database and must be run against the subscription database.
Mrh
•
RPL Buff., 160 MB
Figure 64:
Replay Function and ils Components
The replay can be run in the following three modes:
• Synchronized • Unsynchronized • Replay-only eo&. Po
S ?
Allowa you to:
Run replay proc ••• inq in P.eplay-only 1IIOd.. Run "apl..y p"""' • • • ing in .yncl>ronized lIlOdoo. Get help on thi. _nu . Ib,it the Adal>a. _ n t Replicator Sub. y . _
Figure 65:
Replay Modes
D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_5, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
64
5.1
5 Recovery After a Crash
Replay with a Token
Initiate Replication Replay
From the 'Event Replieation Subsystem' and main menu seleet 'Administration' and from the administration menu select 'R' for initiate replay. In synehronized mode - that is Software AG's recommendation - the replay fune·
tion suspends new Adabas transaetions, reaetivates fil es, subseriptions, destinations that are inaetive and writes all new data to SLOC. lt also synchronizes the data with the new Adabas transaetions when processing is eomplete. ..... A 0 A 6 A 5
16:01:22
EIIl!:N'l' Rl!lPLlCA'l'OR SUßSYSTI!:M . . . . . Adlt.i.ni.tration
Funetion
Coda
D 1
P R 5 T V ?
2010-05-28 M-RPllOO
Dat&ba.a 10 P.rfODl Initial-Sute PLOG Infor.ation lnitiate R40p1ay Sy.tam Funetion. Ta~t Adapte r Global V.lue. Help B"it
Code COIOIIIand
_>
Bnte r -PF1- · -PP2---PF3---PF4-·-PF5---PF6---Pf7--·PF8---PF9---PF10--PFll--PF12--Help .... i t
Figure 66:
Start Replay Using Event Replicator Subsystem
ARrO0147: R40play proe ••• initiated - Tok.n-1582B 16 : 42 : 38 . . . . . A D A 6 A S EIIl!:N'l' Rl!lPLlCA'l'OR SUßSYSTI!:M . . . . . Initiate Raplieation Rapl.y
2010-12-23 M-RP2010
Synehroniz:.d DBID Automated Timeout Fr_ Date/Time To D.te/Time Start D.te/T ime o..tination N _ Li . t Sub.eription N_ Li.t
COIOIIIand
'"
N(Y or N)
'"'--2()10-12-20
+ • D251()09_
()2:0():00
D251010_
0251011_
_>
Bnt.r-PF1---PP2---PFl---PF4---pr5---PF6---Pr7---PFB---PF9---PF10--PF11--PF12--H. lp Z"it Sub S.l W.nu
Figure 67:
Initiate Replay Using Event Replicator Subsystem
5.1 Replay with a Token
65
DBID
Spedfy the database ID of the Adabas database from which you want replicated transactions replayed. Automated
lndicate whether or not you want the replay automated or not. Valid values are Y (perform an automated replay) or N (da not perform an automated replay). The default is N. NOTE: lf you have set the Record PLOG Information (RECORDPLOGINFO) parameter to N (or NO), you caIUlot set the Automated fie ld to Y. lf you specify N for the Automated field, there are manual steps you must perform. Specifically, once you have initiated the replay on this screen, a token is generated that must be used in a manually created ADARPL utility job. In addition, prior to running the ADARPL utility job you might also have to issue
a force-end-of-PLOG request to the Adabas database and wait unHI the resulting PLCOPY job has copied or merged the latest PLOG data set. If you set the Automated Held to Y, none of this manual processing is required . Timeout
Optionally, specify the length of time, in seconds, at which the replay request should time out. The default is 900 seconds. From DatefTime
Spedfy the date and time from which replicated transactions should be replayed . Dates should be specified in YYYY/MM/DD format; times should be specified in HH:MM:SS format. Replay processing will start with transactions in the PLOG that ended at or after this date and time. From dates and times must be earlier than the CUTTent da te and time and earlier than the specified end date and time. To DateITime
Specify the date and time to which replicated transactions should be replayed. Dates should be specified in YYYY/MM/DD format; times should be specified in HH:MM:SS format. Replay processing will stop with transactions in the PLOG that ended before this date and time. End dates and times must be later than the specifi ed start date and time.
66
5 Recovery After a Crash
Start DateJTime The date and time of the PLOG entries that should be used as a starting point for
the replay processing. This date and time are used to identify the PLOG with which to start replay processing. Dates should be specified in YYYY/MM/DD format; times should be specified in HH:MM:S5 format. Replay processing will search the PLOG with this start date and time first for records that match the other replay processing criteria listed on this screen.
A start date and time must be specified if an automated replay is requested. Destination Name List
Tab to these fjelds and type in the names of up to three destinations for replay processing. If you would prefer to seleet the names from a list, place the cursor on one of the three fields and press PF6. If you need to enter more than three destinations and to review the complete list of destinations, type an X in the singlecharacter space for this fjeld and press Enter. A screen appears on which you can maintain the eomplete list of destinations. When the replay request is initiated, transactions will be replayed that were originally destined for the destinations on this list.
lf you want to seleet all destinations for replay processing, tab to the first large spaee for the Destination Name List field and enter an asterisk (.. ). Note that onee you have entered an asterisk in the first field, you can no longer seleet any specifie destinations (errors will oecur if you try). In some versions, the file list is limited, see ehapter 6.4, Problems and Solutions.
Subscription Name List
Tab to these fields and type in the names of up to three subscriptions for replay proeessing. If you would prefer to seleet the names from a list, plaee the cursor on one of the three field s and press PF6. lf you need to enter more than three subscriptions and to review the eomplete list of subscriptions, type an X in the singlecharaeter spaee for this field and press Enter. A screen appears on which you ean maintain the complete list of subscriptions. When the replay request is initiated, transactions will be replayed that were originally solicited by the subscriptions on this list lf you want to seleet all subscriptions for replay processing, tab to the first large space for the Subscription Name List field and enter an asterisk (.. ). Note that onee you have entered an asterisk in the first field, you can no longer seleet any speeific subseriptions (errors will occur if you try).
67
5.2 Replay Without a Token
Up to Event Replicator V6.2, the file list is limited, see chapter 6.4, Problems and Solutions.
Important info: do not use the HALT command for Reptor if data is in SLOG. A Reptor restart will delete this leftover.
5.2
Replay Without a Token
Replay without a Token performs an unsynchronized Replay. All fil es to be replayed must be defined in the job. There are also other parameters that might be necessary, as FROMDATE, FROMTIME, TODATE, and TOTIME. 14: 33: 46 14:33 : 46 14: 33: 47 14:36 : 49
:rAT2000 IZP403l IEPlB9l IEP404l
A 0 A R P L
JOB STREPLAY (JOB59127) SBLZC'l'EO sn STREPLAY - STARTZO - TIM&-14.33.46 PllQEILEM PROGRl\M ATTRIElUTBS ASSlGNKD STREPLAY - BNOZO - TXME_14.36.49
va . l
SKl
0810 _
00039
st...:ted
(ln SAR)
2010-10-20
14:33 : 45
Par_ter.: lUIARPL R&PLAY lUIARPL ~600000K lUIARPL rla$-74
lnitiali"'ation h ... c:lfhak. for Reptor 302 a ucc • •• ful, Tarmin.tion hanc:lfh.ke for Raptor 302 .ucce•• ful
tok. n _ 15045
Total PLOG Block. Read _ 29787 Total "-t>lic.t.d tr .....ction . _ 21285 Kiqhw.ta r mark in r ep1ic.tion pool _ 1634940 T~.tamp of 1 •• t committad tr .....ction • 2010 - 10-15 14:28 50 Total " a p1ic.tad tr ... . . ction • • • nt to hpto" T."qat 10 302 • 21285
A 0 A R P L
Figure 68:
Tarainatad no.,..11y
2010-10-20
14 : 36:47
ADARPL Replay Output without a Token
After the REPLA Y, the source and target file 74 were checked with a self-written checksum program, see appendix, chapter 9.5.7.
68
5 Recovery After a Crash
10-10-20 10/20/2010 14 55: 30 . 0 Started CHltSM014
File FILE-OH DBID
Che ck. um Campari. oft Fil.' : 74 CPU T ....
beorda
OBID be1nq .cc•••• d : ~y
Fi.ld AE
~rom
14:55:30
3!J
Fil. 159 r •• d on wr 2: 1209074'"
39 2386' 4946D1UDl1707A80P526lAA1'17068BB655C6APC 00 : 00 : 00.0 DElID beinq ace ••• e
• Initial-state using the Adabas Event Replicator Subsystem Synchronization will be done automatically. Disadvantage: huge files (>100,000,000) can take too long because the replication process using Adabas commands to replicate • Decompress source file, compress file, re-Ioad target file Synchronization must be done manually Benefit: Adabas utilities are faster than single Adabas commands From the 'Event Replication Subsystem' main menu seJect 'Administration' and from the administration menu T for Perforrn lnitial-State.
69
5.4 Initial-sta te
16:01:22
..... A D A B A S
EVEN'l' REPLlCATOR SUSSYSTEM . . . . .
Adaoini.tration
2()1()-()5-28 M-P'Pll()()
I'uneUon D I
Databa.a ID ~r~orm I:nitial-St..ta PLOG Inf'o"",.Uon I:nitiata Raplay Syatam l'unetiona Tarqet Adaptar Global Valua. Halp Exit
P R
S
T V ?
Coda ...
CommatId
_>
Bntar-PF1---Pr2---PF3---PF4---PF5---PF6---Pr7---PF8---PF~---PF10--Pr11--PF12---
Kalp
Figure70:
Exit
SYSREITOR - Perform Initial-State
16:13 : 15
• •••• A D A B A S
EVEN'l' RBPLICATOR SUSSISTEM •• •••
Parf'or. I:nitial-Stata Initial-Stata N..... Sal
IDBI:D I:Fila I
+0100 1 +01003
Salaetion Critaria
2()10 - ()5 - 28 M-P.P4012
. .. ISl
0
Entar- PF1 --- PF2 --- PF3 --- Pr4 --- PP5 --- PF6 --- PP7 --- PPS --- PF~ --- PP10 -- PF11 -- PP12 ---
Halp
Figure 71:
Exit
Sub
SYSREITOR - Perform Initial-State
Menu
6
Problems and Solutions
OUTing many replication tests problems occurred in the following areas and here are same solutions based on my Qwn experiences. • • • •
Source or Subscription Database Replicator Eogine Target Datahase Utilities
6.1
Source or Subscription Database
6.1.1
Replicalor Buffer (LRPL) Overflow
The Replicator buffer must be defined in the source datahase and in the Replicator Eogine by using the LRPL parameter. An overflow in the source database leads to a halt of the replication process for the file using this buffer at the time of the overflow happened because 00 overfl ow area (SLOG) is available. In general, the same situation happens with a replication overflow buffer in the Replicator Engine, even if an overflow area (SLOG) is defined but without a threshold. See also chapter 6.2, Replicator Engine. 13 : 47 :01 DBJ:D 134
, ,, ,, ,, 'W, ,, ,, 1'001 ". ,, 1'001 ,
Pool /
Qu. u.
Attached autt. r(NAB) C~nd Qu. u . (NC) (LJ"P) Fora. t Pool ,~, Hold Queu. ISN-Liet Tllbl. (LI) 'oq. ~. Tllble (LQ) Ue. " Que u e Unique OE Pool (DUQ) Securi ty (LCI') UQ Pil . Li e t (UQP) ATM Tran e. (:nO) Wo"": Pool (LWP) Red (LRDP)J: ,~) Replic.tion
Figure 72:
2011-01-13 1'AC!1H02
. . . . . A 0 A B A S BASIC SERVICBS . . . . . High W. t. r ~rk .
,
Si" .
1638400 19200 3000000 1134056 5000 20000 312312 60000 10000 96360
I I I
I I
I I
I I
, ,, , ,, I
I \ U.ed I
2152116 6528 2111111840 60088 1072 12784 115500 6208
D.t.
T ......
I
,
I I I I I I I I
13.1 34.0 911.11 5.2 21.4 63.9 36 . 11 10.3
I I I I I I I I
39000 I
4 0 .0 45 .8 0 .0 88.2
I 2011 - 01 - 13 09:51:14 I
, " "" "
1 50 0000
687124 I
5 24288000
4624837116 I
"
I I I I
2011-01-13 2011-01-13 2011 - 01 - 11 2011-01-13 2011 - 01 - 13 2011-01-13 2011 - 01 - 10
08 : 10 : 04 10 :08 20 : 18:43 11 : 24:011 09:48:30 09:411:06 19 : 01 : 18
'"
I I I I I I
, , 2011-01-12 13 : 13 : 14 , ,,
LRPL Overflow
D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_6, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
6 Problems and Solutions
72
Tests in January 2011 with mass updates showed that more than SOOM should be set for the LRPL parameter.
6.1.2
High CPU Usag. 01 Soure. DB
When RPL was switched on in the production database and 9 batch jobs started eru time and the EXCPs of the source database increased dramatkally. The production database used more than double the eru time during replkation. Also, the 9 batch jobs that would normally finish in 2.3 hours now took 13.3 hours. in parallel, the
Based on statistics, these 9 batch jobs update 6 million records per haur update commands per secand.
Of
1.666
The table of Figure 73 dis plays the high eru time occurring in production on 12 Jan 2011 at 19:00 when the 9 batch jobs started and replication was active. Rep1icatioo was turned off on 13 Jan 2011 at 8:00 letting the 9 jobs fin ish before the online time began, JOB
IN TB TIME
PROGRAM
CPU TIME
EXCP TOTL
OB OB OB OB OB OB OB
AOARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN
0:03:45.83 0:03:03.99 0:02:04.08 0:00:15.25 0:42:1 1.47 0:47:04.52 0:41 :43.81
953,021 843,886 629,517 21 ,540 5,053,538 2,823,878 1,940,700
12JAN2011 :15:00: 10.41 12JAN2011:16:00: 10.41 12JAN2011 :17:00: 10.43 12JAN2011:18:00: 10.43 12JAN2011 :19:00: 10.43 12JAN2011 :20:00: 10.43 12JAN2011 :21 :00:10.43
OB OB OB OB
ADARUN ADARUN ADARUN ADARUN
0:41:37.33
2,278,851
0:15:04.88 0: 11 :15.67
3,748,049 5,823,365 2,749,801
13JAN2011 :07:00:10.44 13JAN2011 :08:00:10.44 13JAN2011 :09:00:10.44 13JAN2011 :10:00:10.44
(snip)
SOURCE SOURCE SOURCE SOURCE SOURCE SOURCE SOURCE (snip)
SOURCE SOURCE SOURCE SOURCE
0:06:54.78
(snip)
Figure 73:
High CPU Time - 19:00 Start of 9 Batch Programs
Figure 74 compares the both single and total duration of the nine jobs in December 2010 without replication (2:20 hours) and in January 2011 with replication (13:22 hours).
73
6.1 Sou rce o r 5ubscription Database
24 December 2010 Wlthout Replicatlon Fro m - To
Job
Durat ion
12 January 2011 Wlth Rephcatlon From - To
Durat ion
5031
19:01-21 :00
1:59
19:02 - 07:27
12:25
5032
19:01 - 20:58
1:57
19:02 - 21:38
2:36
5033
19:01-20:57
1:56
19:01-07:11
12:10
5034
19:01-21:06
2:05
19:02 - 07:58
12:56
5035
19:01-21 :21
2:20
19:01 - 08:22
13:21
5036
19:01-21:18
2:17
19:01 - 08:23
13:22
5037
19:01-20:59
1:58
19:01 -07:29
12:28
5038
19:01-20:59
1:58
19:01-07:30
12:29
5039
19:01-21 :00
1:59
19:01 - 07:33
12:32
TOT
19:01 - 21 :21
2:20
19:01 - 08:23
13:22
Figure 74:
Duration of 9 Batch Jobs With and Without Replication
After experiencing such a huge difference when replication was tumed on, the following tests in a different environment (LPAR) under production-like conditions (database, parameters, etc.) were run with only 20% data. Figure 75 shows four different tests with parameters of the source or subscription database and the duration (elapsed time) of the 9 parallel running jobs.
2
YES
40min
3
YES
74min
4
YES
Figure 75:
8
61 min
Duration of 9 Jobs With and Without Replication and Sou rce DB Parameters
Figuee 76 shows the CPU time of the souece oe subscription database with and without replication tumed on . The parameters are displayed in Figure 75.
74
6 Problems and Solutions
(PU Times ~ ~-----------------------
,, +-----
. (PU Time
20
o 1 w/o RPl
Figure 76:
2 w/RPL
3w/RPL-t
4w/RPlH
Duration of 9 Jobs Without and With Replication
The Strobe report (Figure 77) shows that the Adabas module ADANC6, responsi· ble for buffer pool activities, has most activities in the source database (up to 95/0) bu t not in the Replicator Engine. The module ADANC8, responsible for compress and decompress, surprisingly is not shown in the Replicator Engine, even if the decompress process takes place in the Reptor. Adabas Source Database - Strobe report SECTION
.~
. ~~ . ~,
. ~~ .~ . ~~ .~~ .~
. ~~
. SUPII:RVS
Figure 77:
LlNE
PROCEDURE!FUNCTIOK
- ~.
~co
SVC 114
-" _co
~"
~~.
~"
ADAIOS
-"
-"
SrARrING LOCArIOM
PROCEDURE LEMGrB
CPU SOLO
51.U 12.61
ADABAS BUFRit POOL MNGJ:R ADABAS SYSTEM SERVICES
." ."
~~
ADAlIAS THREAD SELECTIOK ADABAS CXPIUISS tlV'l' HNl)I,.R
.u
." ." .'" ." ."
ADASAS USR!BYP EXIT CALL ADABAS 1/0 PROCBSSING ADABAS QUlCUE PROCJ:SSING
ADABAS FORMAt TP.ANSLATOlI SUPERVISOR SJtRVIC!C9
IGCOO2U.
Strohe Report - Adabas Source Database
Adabas Replicator Engine - Strobe report ~= ,~
~
. ~-
-.
SECrIQN
~,
. St1PJ:RVS
IGc:0024A
..~ ~-
-co
~~
.~
~~,
.~
fNC 114
.~-
~~,
. ~-
.NUCLEUS
. ~-
Figure 78:
-"
IBAWSLIt;
-"
PROCE!IIiRII:/n.mC'l'ION
U~
~
~
~,
_co
-"
~.
-"
~~,
PSBSTft
S"tS'lEM SUVl:CJ!.S St1PERVlSOR SERVICES ADABAS USR!HTP DIr CALL ADABAS BurnR POOL MMGER ADABAS 1'IIRZAD s&UCrlON ~
~~
ADABAS ~r rRANSLATOR ADABAS QUEUE PROCBSSING SUSPEND LOCK SZRVl:CZ ADABAS 1/0 PROCBSSING
Strohe Report - Adabas Replicator Engine
SrARrING LOCArIOt/
_m~
LEMGrB
,~
~w
21.27 13.U !l.O!l 9.18 1.21 .00 3.64 2 .13 2.13 2 . 13
6.1 Sou rce o r Subscription Database
75
LBP and LFIOP at Source and Replicator Engine Buffer Usaga Durlng Repllcallon
• Da ' a F/ow·
OB 13. (PRODI Soure . I S"b sc rlp1lon OB
L
ru
L_
I
RH 8uff. , RPl _5 H,2U ,000 U .....d
lBP - 10 ,000,000
I
I
OB 312 jR.plorl R. pllc.1or Eniline
"
I,
'---
I
J
" ~
,~'''"
LFIOP _ gOO ,OOO _ 25 .. 01 LBP
8
1];. 000 .' "u ,h . .
~u''' '
000 .' "u ,h.,
~ . ""
I
..... , TU .. .
I
..!J. lBP - UOO .OOO
ry.
1/0 Bufl .. lFIOP - 25 .. ollBP
I
RPl8uH. , RPl _5 2. ,2 U,OOO 13"u u d
J
I ~u". '
$lOG .... 30
Dieter Storr - 14 Jan 2011
Figure 79:
Buffer Usage Between Source OB and Reptor
The Adabas buffer pool (LBP) on the Replicator Engine was set to a very low value of 3.6M compared with 70M on the Source database. lf updates on the Replicator Engine cannot be passed in time from its RPL buffer to the target database then they must be stored into its subscription log (SLOG), an Adabas file. And these updates must be passed through Reptor's buffer pool (LBP) and l/O pool (LFIOP). lf the data to be replicated cannot be passed fast enough from the source database to the Reptor then the RPL buffer of the source overflows. It is also a CPU overhead on the source database to ask very often the Reptor whether to be ready to hand over the next transaction.
Recommendation: LBP size of both source database and Replicator Engine should be the same, at least 150M.
76
6 Problems and Solutions
•• •• • 11. D 11. B 11. S
12:11:21 DBID 34
Ar..
-------- - ----- Pool a --------- - ----
Sort
(~S).
Int. u ••" Bu~~. r Suff. " Pool Format pool ISR Liat Tabl.
49920
(LU). 500000 .. 150023680
(~P)
5500000 (LI) .. 5000 (LQ). 20000
(Ln).
S~. Cmd. Table Work 5'001 Attachad Buttar Sec:urity Pool
(loWP) .. 1500000 (NAß). 400 (l.ci') .. 10000 (LDI!UQP). 60000
UQ-DB Pool rluah 1/0 Pool (L .. rop) .. 13000000 Br". R
,
00
30 HS NO NO
2011-02-24 PACPD02
------------- Qu.ue. -----------------(NC) 100 C....... nd Qu.u. Hold Qu. u . (NH) .. 40500 U•• r Qu. u. (NU) •• 1000
------------ TLme Window. ------------300 3600 3600 3600 900 3600 180 3600 ISO 3600
Tran •• ction T~ (TT) Me>< T.,an •• ction T...... (MXTT) Nonactivity JU:C-u •• ., (TNM) Non.cti ... ity ET- U••., (TNAE) Nonactivity zxu-u •• ., (TNAX) Me>< Nonacti ... ity T ...... (Mn'NA) T~ Limit Sx-Cmda (TLSCMD) Me>< T~ t"o., Sx-Cmd.. (MXTSX) C _ n d T~ (CT) SmS60 Inte.,...a1 UNTNAS)
-------- u ••., Specitic Limit. --------Hold Qu. u . L.im.it (NJ:SNHQ) 10000 CID. pe., U•• " (NQCID) 50 ISN pe" TBl: El_nt(NSISN) 51 ------------ Butt"• ., Pool -------------But"t."t"lu. h Du.,. (Tn.t1SH) 1 1'• .,.11.1 LFIOp 1/0 (~IO) .. B Aaync. by Vo1-S." (AS!TVS) HS ------------ Replic.tion -------------P.eplic.t:ion (REPLICATION) .. . l!:S
118000 00
,
43000
'"
---------- p.,ot. ction Loq<Jinq PLOG ".qui.,.d (PLOGRQ) 00 DUAL PLOG Siu (DUALPLS) 13500 (DUALPlD) 3390 DUAL PLOG D.... ic. NPLOG ...•...•..••..•...•..••..• 0 - ----------- Oth • ., S • .,...ice. ------------Tri'1\l'.r. end Procedure. (SPT) NO D. lte S ..... V.cility (DSF) NO C.ch. r .. cility (CACHE) NO Tran •• ction Man.q• ., (ATM) NO TCP/IP SUppOrt (TCPJ:P) NO Ext. I:.,.,or Reco .... ry (SMGT) NO 2 Ph.... C..-it Suppo.,t (DTp) NO Revi. w Support (REVIInf) NO
High CPU Time - Adabas Source Database - Used Parameters
The I/O buffer (LFIOP) is responsibIe to do asynchronous write I/Os to disk after fil led with blocks from the buffer pool (LBP), caUed buffer flush. It should be smaller than 25% of LBP minimizing Reptor's wait time to do the next buHer flush.
6.1 Sou rce o r Subscription Database
77
LRPL at Source and Replicator Engine
The RPL buffer of both Source database and Replicator Engine should have at least the same values. lf the RPL buffer of the Replicator Engine is also used to store data into the SLOG then the RPL buffer of the Replicator Engine must hold also the data from the SLOG to the target database via the buffer pool (LBP) and the replication pool (LRPL). Recommendation: LRPL=700M ADADBS DSBI=OFF 00 not collect before images of data storage for replication during the update of a
record on a file. OSBl=QN is only necessary for filtering data in a user exil. The parameters controlling the replication on the source or subscription datahase per file should be: JlDJIDBS REPLlCATION PlLE-006,0N, TARG.I!T-312 ,DSBlooOPP JlDJIDBS REPLlCATION I"ILE_ 001 ,ON, TARG.I!T-312 ,DSBlooOl"I" JlDJIDBS REPLlCATION PlLE-008 ,ON, TARG.I!T-312 ,DSBlooOPP .~.
Figure 81:
ADADBS REPLICATION and DSBI=OFF
ADARUN ASYTVS=YES .nd FMXIO=8 (m.x 16)
The FMXIO parameter sets the limit on the number of 1/0 operations that can be started in parallel by LFIOP flush processing. The value of FMXIQ must be adjusted after discussions with the system programmer if parallel access volumes (PA V) are available and the SLOG file is used causing replication data to be written and read from the SLOG file. Recommendation: both source database and replicator engine should be set to ASYTVS=YES and FMX IO>l, depends how much parallel 1/0 activities can be handled for PAVs. Log Input Transaction f LOGINPUTTRANS
This parameter defines the threshold of Reptor' s RPL buffer before using the Adabas file SLOC. It was reduced from 75% to 50%, regarding a former recommendation from Adabas Technical Support to avoid Reptor's lock in stress situations. Now, SAG recommends setting the threshold to 70%.
78
6 Problems and Solutions
AlU'O0045: o.ta"lt info "p:t.~d ..... A D A B A S 08:22:42
EVBNT RBPLlCATOP. SUBSrSTZH . . . . . Global Valu..
,
Subta a k .
Mo. ......._
;===:" !:~:~y ~.~ . ~~ . ~~~~~~ .. ~
OUtput Siz • . . , •..••..•• -
Enti"eX Brok." Stilb N _ Input Requ •• t M.s_ Compla t1cm Leva l . . Compl. tion t. .... l
..• . .••..• . ..
,,
Rooq\MIat b e . i ....." LevMIat Zr"or Loo ....1 0 St.tu. Requ. •• t Le ... l ... . .. . I-Stat. Start Raqu •• t Loov.l
I-State Compl.tion Lev.l Retr ......... t Req\MIat r.. ....1
,,, ,
Subta . k ~tiv.tion Wait Open D•• tination. at Start C-nd - > Znt.r-PF1-- -PF2---PF3---PFf ---PF5---PF6-- -PF7---PFB---PF9---PF10--PFll- -PF12--Ha lp
Figure 82:
High CPU Time - Event Replicator Subsystem - Global Parameters
Use the Log Input Transaetion field to specify whether or not the Event Replicator should use its SLOG system file as a temporary storage loeation for incoming compressed replication transactions, before they are queued for processing. Once transaetions have been written to the SLOG system file, the Event Replicator Server processes them using a throttling mechanism so that only a Iimited amount of Event Replicator Server replication pool space is used at a time. Depends on the RPLPARMS parameter setting (FILE, PARMS, BOTH), LOGINPUTIRANS can be read from DDPRINT and/or from the system file. More than one replicator engine (Reptor)
Based on the number of files to be replicated and the number of files in one Adabas transaction, it is recommended to separate heavy updated files for Reptor A and B. This can be valid for 100 and 200 replieated files or transaetions containing 15 and 20 updates. The Reptor collects all updates for one transaction before it sends to the target database. Source DB stopped to replieate all files If you use UTILONLY=YES before you apply mass replication parameters in batch it will stop all replications.
6.1 Sou rce o r Subscri ption Database
lUIlUIBS ADADBS lUIADBS lUIlUIBS
REPLICATION REPLlCATlON REPLICATION RBPLlCATION
79
FlLE-OOl ,N:lDlrY ,DEACTlVATII:, TARGET-312 FlLE_002,N:lDlFT , DBACTlVATB , TARGET-312 FlLE-006,N:lDIFY,DBACTlVATB,TARGET-312 FlLE-012 ,N:lDlrY ,DBACTlVATB, TARGET-312
lUIADBS RBPLlCATION FlLE-201 ,N:lDlrY ,DBACTlVATB, TARGET-312
Figure 83:
ADADBS REPLICA TION DEACfIV ATE
Software AG recommends using the SY AOS function 'Modify file parameters' to activate w/o stopping the users.
6.1.3
ADAFRK - Transactions Turned Off
DDPRINT showed the message ADAFRK dbid transaction turned off. 14: 39: 32 14: 39 : 32 15:58:05 15: 58: 05
Figure 84:
ADAJ'RJ 00302 U... 01' SLOG I'or dAt
Figure 92:
Manual File Change Procedure - Open the Destination
6.2
Replicator Engine (Reptor)
6.2.1
NAB Overflow
The Replicator Engine uses for each update command (store, update, delete) in one transaction aseparate attached buffer (AB). It co1lects and holds a1l update commands until the transaction receives an end-of-transaction (ET) command. It sends then all Adabas commands of this transaction to the target database .
_.,Ip"', ......., •• 11'
I
NII.wt l
Figure 93:
I
ADAP18
.- -_. --......, -
-U211
1 I
NII.0421
1 I NIl.·... I
11so Itoutoors
~-
"nNo, 1.......
~
"
.1 ...". ~ ~
1-
~
NAB Overflow on Reptor
Al c:-.d to 0810
Figure 94:
."
_
300 FNR
NAB Overfl ow Message
18 RSP
2 5 4 aubcode
3 ISN
D8 300 Destination
er_rg,t)
87
6.2 Replica lor Engine (Replor)
RSP 254 = An internal error occurred during processing of an attached buffer (buffer overflow) Subcode 3 = CQEFLAG is not equal to CQEFAß plus CQEFW16 Solution: Replicator: NAß 420 is too small and should be increased . Each update command (store, update, delete) in one transaction (ET) needs one attached buffer. The formula in the manual is not correct. It must be: NAß = 41 x 10 x number of source databases (e. g. 1) x number of updates per transaction (e. g. 15) NAß=41 x 10x1 x 15=6,150 The high-water-mark of the attached buffer (NAß) is observed with the AQS command.
- -Tb •
• "tu.l ~or
biqh - ... t . r - ... rk . (. xcapt bu~f . rpool)
th. major pool .
~,-
HIGH-~TER-~
-----------------------------------------------
-""'" DUO -POOL '" -""'" -""'" "'Q -""'" lU'L -POOL ,e -""'" TBI -POOL ~
TB8 -POOL
'"
-""'"
U-POOL XID - POOL
Figure 95:
6.2.2
'"
23040 w_ 100000 300000 560056 ~ ~~ 157286400 ~~ 10000 12000 20000 4000 ~ 4000 ~ 500000
~ -
"'"' --
-
,,~
,, ", ,, ,, "" "" , , '" '", , , ,, ", "" '" , ,• "" ,, , " , ", "" "
1715456 1920
on
29888
...
149875480
17248 3168 53112
NAB Overflow High Water Mark
Replicalor Buffer (LRPL) Overflow
The Replicator buffer must be defined in the source database and in the RepUcator Engine by using the LRPL parameter. An overflow in the source database leads to a halt of the replication process because no overflow area is available.
In general, the same situation happens with a replication overflow buffer in the Replicator Engine, even if an overflow area (SLOG) is defined . The trick is that you have to set a threshold in Reptor's RPL buffer. If the threshold is reached, for example 75%, all incoming transactions will be stored into the SLOG area.
88
6 Problems and Solutions
After new tests in Oetober 2010 with mass updates the parameter LRPL is set in both source database and Replicator Engine to 600MB. The high-water-marks
._.
show now for the souece database 80% used and for the Reptor 74% used .
I DI.n ......-
--013.2
(_ _I
I
LIt'l.
,.mp,·,Zr 'IIMB
r.cords
De-eomp ISled
LIt'L_
''''''1
I
.
\
\
SLOG • AdMI •• nie
\
.... it .. Cri
r.cords
--
-
I
ErotIR Noot· ~.
ARF •. 3.t.t, ..expanded LRPL rrom 160M B 10 30 0MB
-S .t RPl limit 10 75 %
Globe l V.lu.5Ilog Input Trans Belion 07 5
.1 ••• 11 CrI
Figure 96:
-SlOG will b. " lied lIfter 75 % SLOG 111 un d
LRPL Overfl ow Reptor
The high-water mark of the attached buffer (NAß) is observed with the AQS
command. Tb• •etual ~or
h1qh- ... t . r-.arka th. _jor pool, (.xcept bu.f:f:.rpool)
-
ADARUN PARM
U
'" "" "
'.
~L
"m
n, UQ
-"",L -"","
-"",L -"",L -"",L
-"",' -"",L
-"","
- "",L -"",L
00" - POOL WORK-PQOL XID - POOL
Figure 97:
.c "
'"
23040 W~ 100000 L>~ 300000 560056 ~ " ~~ 1572B6400 ~~ 10000 12000 L' " 20000 4000 ~ " 4000 ~ " L_ 500000
"' " ,,~
.., ""
HIGH-~TER-~
1115456
1920
, .,.," ., '" " .,., '" ,•• .,., .,., "'" on
211BBB
...
0 0
14I1B754BO 0
0
17248 3168 53112
0
LRPL Overflow Reptor H igh Water Mark
89
6.2 Replica lor Engine (Replor)
6.2.3
SLOG Overflow
The overflow area of Reptor's replication buffer (SLOG = subscription or source log) should be defined with sufficient space to hold all transactions from the source database for a couple of days in case the network or the target database have problems and the destination is c1osed. 1I is recommended 10 use an entire 3390-disk for SlOG 15:01 : 52 OBIO 302
•••••••••••••••• *••• *••• *•• ** •• *
. . . . . A 0 A B A S BASIC S&RVICBS . . . . . - Oi.p1ay I'il . Layout ST-SLOG
bco,,~ loadood . . TOP ISS •..•...•..•• Ha" ISS . xpac:Ud Min~ ISS ...•.. Siz. o f! ISS . . Rumba., of UpdaU. ISN bu•• q. S~ce Rau.aqe ADAM l'i1. Ciph• .,ed. ril. Coupl.d l'i1 •• Blk p"r OS Itxt. nt Blk p"r UI Itxt ... t Blk p"r NI Ext. nt F"• • •pa". a ... ailab1a
I D. ... LiI I 'l'ypoo TyI
,
,
2010-09-24 PtlRI'042
35625 35532 5000655
2010 - 01 - 25 13 : 55:09 Oau loadad. Dau of 1a.t updau 2010-09-24 15 : 01:08 Max Comp" bc Lnq1:h 5060 "".o/OaU Pa cldinq •..• n/n
,
Hiqh •• t Inde" La .... l
4 ByU. 529H
,.. ,..
RPLUPDATBONLY . No USZRISN No MIXD$OIV ••••• No Sp... nad. "." . . No iteplic.tion .• No uni ....".al Encodinq
No No
N=. , o o
f!o" fil .
,
•••
Ind>< Comp
No
PGlCP.EFRIISH
,
NOACEXTENSION .. No MO/PI! indic •• NO
,..
Total Chanqed Blk. .. . 31102 Multi CH.nt I'il• . .. . 0 a"t.nt.: At l.a. t 134 B"tant.
Space allocaud I Bloch I Cy1& . I
FrOlll. RJIBN
To RABN
I unu •• d I BLOCJtS
/
Cy 1& •
, ------------------------1-----------------, ,, ", , , ,
----I--------I-------------------I 5891
ASSOI 3390 ACI ASSOI 3390 UII ASSOI 3390 NIl
,
,
291000
'"
1100 I
~.~
3390 OSI
995700
6638 I
Figure 98:
1780 304671 1677
7676 I 304816 I 304616 I
0 194 296556
1098
996250 I
892920
5952
Adabas Reporl SLOG file
Check the follo wing parameters if SLOG is used as an overflow for the RPL buffer on Reptor
• • • • • •
• •
ASYTVS FMXIO LBP LDEUQP LFIOP LP NH N ISNHQ
Async Buffer Flush Based on VolSer# Max Number of Parallel I/Os by LFiOP Flush Proc Buffer Pool Unique Descriptor Pool Asynchronous Flush Pool Work Part 1 Number of Hold Queue Elements Number of ISNs in Hold Queue for Users
90
6 Problems and Solutions
6.2.4
RSP 153 from the Target OB
According Messages and Codes V6.2 SP2, Adabas can only handle one call at a time. Adabas response code 153 is reported by the Replicator Engine in DDPRINT: 11:07:34 ADArS4 00302 Replication • .,ror: Adabaa deatination 0251011 11: 07: 34 AOAJ'54 00302 Soure. DBID 39 rm:t 11, Ta"qat Dsto 251 FNR 11 AD.U'18 N2 caod to DBID 251 rm:t 11 RSP 153 . utx:ode 16... 8 18N 75423 13:00:04 AD.U'54 00302 a.plieation a.,ror: Adab •• deatination D251044
13; 00: 04 ADl\.l'S4 00302 Soure. DBID N2 c:md. to DBIO 251 I'NR
ADAJ'18
39 FNR 44 RSP
4', Ta"qat DBIO 153
. ub~
251 FNR
16448 ISN
ADAF54 2010-11-22 13:00 :04 Replication • .,ror: Adabaa deatination 0251044 ADAJ'54
Soure. DBID
AllJI.F18
N2 caod to DBID
39 I'NII.
251 FNR
44, Tarqat DBID
... RSP
251 PNR
...
522015
44
153 aubcode 16"'8 ISN
522016
ADAP54 2010 - 11- 22 13:00 :04 Rap11cation • .,ror: Adabaa ~atination 0251044 ADA:f'54 Soun::. OElIO 39 1"NR 44, Tal
Check_ ....... do not _tch
««
Figure 117:
Reptor Under Stress - User Gone Message
6.3 Target Database
105
SAG Tech Support recommended setting the threshold of RPL buffer to 50%. After a high CPU time and a slow handshake between the subscription database and Reptor, Becky Albin from Software AG recommended setting it back to 70%.
6.2.18 Replicalion wilh wrong FOT/GFB lUIAB'7F 00312 Al d.compr ••• ion re.pen. .
Figure 118:
55 .ubc:ode
0 . ub.cr iption S13.017
Response Code 55 - Wrong FDT/GFB
A decompression fa ilure occurred d uring the subscription phase of the Event Replicator for Adabas processing. File changes on the subscription database was not applied in Reptor's global format buffer (GFB). The forma t buffer for the after image (AI) does not fit. Solution :
Oelete the old format and create the new one from Predict.
6.3
Target Database
6.3.1
RSP 52 from the Target OB
FOT on source and target are different. The rightmost two bytes of the ADD2 field contain the sub-code, the leftmost two bytes the name of the field. This problem can happen during program migrations with file changes and the changes are not done at the target database. Adabas on Windows does not automatically replicate file changes. A special customized procedure must be developed to synchronize file changes on the source and target database. See also chapter 6.1.5, ' File Changes Won' t Replicate' . The following error messages (ADAF1 8) are collected from Reptor's active DDPRINT and sent by e-mail to the DBA group to be checked. (.nip) lUIAI"18 lUIAI"18 lUIAI"18
N2 CIDd to DBID N2 CIDd to DBID N2 CIDd to DBID ~~, N2 CIDd to DBID lUIAr18 N2 CIDd to DB:lD lUIAI"18 N2 CIDd to DBID FNR!RSP/CNT 116 52 7513
191 191 191 191 191 191
FtIP.
rNR rNR rNR rNR rNR
116 116 116 116 116 116
RSP RSP RSP RSP RSP RSP
"" "" ""
.uboode a ubc:ode auboode a ubc:ode .uboode a ubc:ode
•••••••••••••••••••••••••••••••••••••••••••••••••••••• •••••••••••••••••••••••••••••• ••••••••••••••••••••••••
Pi•••• ch.ck th. AOAßAS r •• pon •• code.!
Figure 119:
Response Code 52 Showed in Reptor
.9602 .9602 .9602 . 9602 U602 . 9602
". ". ". ". ". ".
145861 145863 145871 145892 14590. 145907
106
6.3.2
6 Problems and Solutions
RSP 176 from the Target OB
The inverted list in the ASSO was corrupted for file 75 in the target database 25l. This was probably a sign of later problem with the HP disk arrays - two raid-5 disk arrays failed at the same time. DDPRINT of Reptor: (an1p) 11:38:08 ~54 00302 S-p1ieation arror: Adabaa daatination D251074 11:38:08 ADl\.F54 00302 Soure. DBID 39 FNR 74, Tarqat DBID 251 FNR 74 ~18 N2 c:md. to DBID 251 FNR 74 RSP 176 . ub~ 35841 18N 35320183 ~54 2010-10-15 11:38 :08 Raplieation .rror: Adaba. da.tination D251074 ~54 Soure. DBID 3S1 PNR 74, Tarq.t DBID 251 PNR 74 ADA:I'18 N2 c:md. to DBI:D 251 FNR 74 RSP 176.ubcoda 35841 ISN 35320190 ADA:I'54 2010 - 10 - 15 11:38 :08 S-plieation . rror: Adabaa daatination D251074 (anip)
Figure 120:
Response Code 176 Showed in Reptor
It was also unusual that the target database ended abnormally during the replication process.
Solution: Unload with ISN and load to re-build the corrupted inverted list.
6.3.3
RSP 98 followed by RSP 113
The target database rejects a store command with ISN 1 (N2) in empty file 80. The following delete command (EI) with ISN 1 results in ISN-I-not-found message. The rest of stores and deletes with ISNs 2-35 ended with RSP Q. Response Code 98 could not be repeated, as requested from Tech Support of SAG (#5037205). The unique super-descriptor contains of an alphanumeric value and ten asterisks (""), for example 'RISTMOOOOO10183 ......·..·•........'. In former Adabas versions for the mainframe were problems with its unique descriptor pool, defined by LDEUPQ. There is not such parameter in Adabas for Windows but maybe the internal DEUQP has not been cleared or there are problems with the asterisks. From SAG's messages and Codes: 'Uniqueness violation of unique descriptor detected during store/update if subtransactions are not activated, or otherwise at end of sub-transaction. The third and fourth bytes of the Additions 2 field contain the name of the descriptor which caused the uniqueness conflict:
6.3 Target Database
107
CMDRSP
FILE
o 00000000I... o 00000068111... o 00000000I ...
eo eo
2203960 07.JAN·2011 02 :53:27
0
2203961 07.JAN·2011 02 :53:27
0
ADAN5S .$ . .... E1 ADAN5S .$. ____ N2
2203962 07.JAN·2011 02 :53:27
0
ADAN5S
2203969 07.JAN·2011 02 :53:27
0
ADAN5S .$. .... N2 98 4F4100001 .. AO
2203970 07.JAN·2011 02 :53:27
0
ADAN5S .$ . .... E1
2203971 07.JAN·2011 02 :53:27
0
ADAN5S
2203979 07.JAN·2011 02 :53:27
0
ADAN5S
2203980 07.JAN·2011 02 :53:27
0
ADAN5S .$. .... N2
2203981 07.JAN·2011 02 :53:27
0
ADAN5S .$. ... , ET
2204291 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... E1
2204292 07.JAN·2011 02 :53:35
0
ADAN5S
2204293 07.JAN·2011 02 :53:35
0
ADANSS .$ . .... ET
2204301 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... E1
2204302 07.JAN·2011 02 :53:35
0
2204303 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... N2 ADAN5S .$. ____ ET
2204310 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... N2 98 4F4100001 .. AO
2204311 07.JAN·2011 02 :53:35
0
ADAN5S
2204312 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... ET
2204320 07.JAN·2011 02 :53:35
0
ADAN5S
2204321 07.JAN·2011 02 :53:35
0
ADAN5S .$ . .... N2
2204322 07.JAN·2011 02 :53:35
0
ADANSS
$ .... ET
.•.
.... ET
o 00000000I.,, o 00000000I ...
0
R
eo eo 0
(ether files)
eo eo
$ .... E 1 113 00000000I...
(snip)
$ .... N2
o 00000068Ih... o 00000000I ...
ISN
"
" " ,
+
+
2
0
o 00000000I ... o 00000068Ih... o 00000000I ...
eo eo
o 00000000I... o 00000068I> o 00000000I ___
eo eo
0
33
"
(o1her files)
0
" "
eo eo
35
(snip)
.•.
.... E1
o 00000000I... o 00000000I ...
R
0
(snip)
$ . .... E 1 113 00000000I...
.•.
.... ET
o 00000068I>. o 00000000I ...
+
eo eo
,+ 2
0
(snip)
Figure 121:
Response Code 98 and 113 Showed in Reptor
Solution:
• • •
Define aB files in the target databases to be replicated with REUSEISN and setup the MAXISN with the same values as in the source database Start the initial-state process RSP 98 was no longer observed during the replication process.
6.3.4
RSP 113 on several files
The specified ISN was invalid because of • N2 command was issued with ISN equal to 0 or larger than the MAXISN in eHect for the file; Adabas on Windows surprisingly adds only 1 block as a secondary allocation (extent) . • N2 command was issued and the specified ISN was assigned to another record in the file;
6 Problems and Solutions
108
•
L1/L4, EI, Al or 51/52/54 (with FB) command was issued for a non-existent ISN;
• L3/L6 command faund an lSN in the index which did not exist in the Address Converter; • Replication definitions: source and target DB/FNR were the same, for example from 06134 FNRl54 to D6134 FNRl 54 instead of to 0625] FNRl54. There is no check in SYSREPTOR to avoid this. • Files on the target database were not defined with REUSEISN. Later during the replication process and after closing the destinations, files were correctly defined with REUSEISN. All files should be defined correctly before the replication process begins. 14: 37: 03 ADAJ'54 00312 Sour". DBID 134 I'NR 122, T."q. t DBIO 134 FNR 122 N2 caod to DBID 134 I'NR 122 RSP 113 .ubcode ISN 12440974 A1JUCV n.. ".cord to b,>,or: Adaba. da . tination D251046 AnAP5 4 Soure. ODlD 134 mR 46, Tarq.t OBto 251 FNR 46 lUIAr18 Al cmd to DBZD 251 rNR 46 RSI' 162 aubcoda 17473 I8N 6023043 ADAF5 4 2011-01-07 21:05:18 Rap1ication . rror: Adaba. da.tination 0251046 ( . n ip)
Figure 124:
Response Code 162 Showed in Reptor from Target DB
RSP 162 means that no additional space was available for the Adabas buffer pool. Increase the value of the LBP parameter. Be 60 Files
Replay with ERROR·139
REPLAY received ERROR-139, ADARPL has exceeded the replication pool size allocated via the LRPL parameter. 80th RPL of the target database and the Reptor are only 2% used, according to the HWM . The utility ADARPL has its own RPL buffer. lncrease the default value of LRPL from 100,000 to 300,OOOK. Tests with many files to be replayed showed that a value of 600,000K is better. J:RRQR-139, ADARPL raplic.tion pool o .... r~low Incr•••• LRPL p.r_e.r .nd r .run th. job.
Figure 131:
ADARPL Replay ERRÜR·139
6.4 Utilities
6.4.3
113
Replay with ERROR·140 RSP 254
REPLAY received ERROR 140 with response code 254 and sub code 6 at loca Hon 1. 4
4
A DAR P L: Brror occurred durinq . .. ec:u tion : ERROR- 140 , Une~cted REPTOR r •• pon •• 254 .ubeode 6 loc.tion 1 2010-05-03 A DAR P L T.noin.ted
Figure 132:
4
17: 39 : 25
ADARPL Replay ERROR-140 RSP 254
The attached buffers (NAß) can be too small defined. ßased on the high-watermarks of the source database and the Replicator Engine, NAß is only up to 20% used. The Replication Utility Manual Version 3.3.1 does not describe a related parameter to the attached buffers. Subcode 6 means that the low-order 6 bytes of CQECKSUM (checksum of command queue element) do not equal UBCKSUM (checksum of user block check). This information seems to be not very helpful for a DBA. It also can be that the CT parameter limit was exceeded (tested with CT=180), the
nuc1eus terminates the user, genera ted the equivaJent of a BT command intemalIy, released the CQE and attached buffer space, and issued ADAM93 or similar message. This message was not detected during this ERROR-l40. No solution or explanation has been found to solve this problem.
6.4.4
Replay with ERROR·1 40 RSP 131 Subcode 70
The following REPLAY was running without a TOKEN: A DAR P L
va . 1
SKl
DBID _ 00134
Sarted
2010-12-23
12 :3B: 2B
~L
LRP L- 600000~
~
FILBS-006,007,00B,009,010,011 ,012 ,013 ,014 ,015,016 ,017 ,01B ,019 ,0 20 FILBSa021,022 ,023,0 24,025 ,0 26 ,02 7,02B,029,030,031 ,032,0 33 ,0 34 , 035 FILBSa036,037,03B,039,040,0 41 ,042,0 43 , 0 44,045,046,04 7 ,04B,049, 050 FILZSa051,052,053,054,055,056,057,058,059,060,061,062,063,064,065
~ ~ ~
A DAR P L : Error occur red dur1ng ....cution: ERROR-140 , Unaxpected RBPTOR r •• pon •• 131 .ubeode 70 loc.tion 2 A DAR P L T.noinatad 2010 - 12- 23
Figure 133:
ADARPL Replay ERROR-140 RSP 131 SC 70
12 :38 : 29
6 Problems and Solutions
114
RSP 131 has up to 111 subcodes bu t suhcode 70 is not found in Messages and Codes V3.3.2, nucleus response codes. Subcode 69 means th at a replay fo r the same DBID/FNR is already running.
ADARPL was running with the following version: Dat. 2007-01-13, V.r.ion 3.1, SM 1, B••• AZ311000 Zap. AZlllOOl AZlll008 AZ311011 AZ311054 AZ311011 AZ311073 AZ)11074 AZ3110B4
Figure 134:
6.4.5
ADARPL Version and Zap Level
Replay with ERROR·140 RSP 131 Subcode 61
Figure 135:
ADARPL Replay ERROR-140 RSP 131 SC 61
Subcode 61 means that an invalid request sent to Event Replicator Server. RBL is insu fficient, no FB provided on lnit handsh ake or other such error. Notify your Soft ware AG technical support representative.
Solution: Just restart the job.
6.4.6
Replay with ERROR·146 - loken
Taken 19655 was created. ARrO0147 : Rep1ay proce •• initiated - Token_ 19655 . . . . . A D A B A S EVEN'r UPLICATOR SUBSYSTEM • • ••• 12:12:34 lnitiate Rep1icatiot> Rep1ay
2010-12-23 H-RP2010
Synchronized OB1:D . . . . . . . . . . . . . . . . .. Automated Timeout , •..••..••..•..
_134
I'rom Date/Time .• . To Date/Time .. Start Date/Time • .
2010-12-20 2010-12-23 2010-12-20
D•• tination Name Li. t . Sub.cription Name Li.t
A D A Il P L
Figure 136:
VB. 1
SM1
DBID _ 00134
l'
(1' 010 N)
'00 02:00 : 13 11:55:00 02:00 : 13
+ •• • _ __
St..rted
ADARPL Replay - Create Token
2010-12-23
12:31 : 57
6.4 Utilities
115
The Token 19655 was not found. There are no replay processes active with the token number listen in the message.
ADARPL RJ:PLAI TQUN_19655 ADARPL RPLTARGETID_ 312 ADARPL ~600000~ l\. D l\. R P 10 :
BRROR-148,
Brror occurr..:! durinq ....cution:
~
19655 not tound in Reptor.
Inv •• tiqate th. cau • • ot th • • rror. eorr.ct i t an
2010 -1 2 - 23
Figure 137:
12 :31 : 5B
ADARPL Replay ERROR-148
Solution :
Deleted the TOKEN, created a new one and started the replay with the new TOKEN number.
Reason: The field 'Automated' (Figure 136: ADARPL Replay - Create Token) should be marked with N and not Y.
6.4.7
Replay Abended with SOC4
Created token 15825 with 183 destinations and subscriptions. The started ADARPL utility ended abnormally with error code SOC4 and reason code 10. This is normally caused because of protection exception, incorrectly linked modules, invalid pointer value, wrongly concatenated Adabas library, etc. l\. D l\. R P 10
va.l
SMl
DBID _ 00134
Stert..:!
2010-12-23
14 :51 :2'
ADARPL REPLAI TQUN_15B25 ADARPL RPLTARGZTID-312 ADARPL LRP~600000K
14: 51: 51 14:51:51 14:51:51 14:51:51
Figure 138:
END 01" SYMPTOM DUMP IZI"4501 STRZPLAl\. STl RZPLAI - ABZND-SOC4 UOOOO RZASON-00000010 TlME_14 . 51.51 IZI"4041 STRZPLAl\. - ZNDBD - TIME_14 51.51
ADARPL Replay SOC4
6 Problems and Solutions
116
Restarted the replay job with the latest Adabas steplib and with a new Taken 15826 but received the same SOC4. The dump can be viewed in appendix, chapter 8.7.
Solution: Cleaned up 15826. 1 6:24:35 Replicator 312
. . . . . A D A B 11. S
BASIC
SERVICES • • • ••
2010-12-23 PRPT002
Replicator Hanaq_nt
A D P
At1 ... ate/&o.ct/~n/"lo •• Ci . pl ..}' Jteptor de f inition. D1aplay Rooptor atatiati"a
H L P
' . "form. Rl'LCl •• nup
,."torm.
RPl.Chack
P."form RPLRefr •• h
parameter aubayatam ?
Salp Exit
Code D.~ ••
ERPTOOl
C_nd
10 .. 312
(RPLPB-DATA-BASE)
_ >
PPl----- Pt2------ PPl------ PPt------ PP6----- PI7----- PPS----- PP12--- -Pile Sarv S"it
flalp
Figure 139:
ADARPL Replay SOC4 - RPL Cleanup
Solution:
Created multiple TOKENs with only a couple of files. ARPO0141 , RIopl.y p"oe ••• initiatad - Tokan-15828 16:"2:38 . h ._ A 0 A 6 A S EIIl!:N'l' RJ:PLlCA'l'OR SUßSYSTI!:M H ••• In1tiate bp11c:at1on bp1ay
2010-12-23 N- RP2010
Sync:hroniz:ad OBIO Au_ted T>.m.out
_13" N(YorN)
Fr.". Oate/T>.m. To Oate /T ~ Start Oate/T>.m.
2010-12-20
Oeatination N _ Li.t Sub. c:ription N_ L1. t
'''---
+
._-
0251009_
02:00:00
0251010_
0251011_
Command _ > Enter - PP1 --- PF.'2 --- PP3 --- PP .. --- PP5--- PP6 --- PF1 --- PP8 ---PP9 --- PP10 --PP11 -- PF12 ---
Figure 140:
ADARPL Replay SOC4 - Create Multiple Token
6.4 Utilities
117
Solution: Ran multiple Replay jobs in parallel, each with different TOKEN, and with the same shared PLOGs.
6.4.8
Replay with Response 131 Subcode 61
It seems that you cannot define more destinations than fit on one selection page (PF6).
JU!.F(J0156 : Re.pon •• 131 S~ 61 Rec.iv.d trOll!. nop1ic .. tor 1 7: 56:56 ..... A 0 A B A S EVENT RZPLlCATOR SUBSYSTEM ..... lniti.t. ~1ic.tion ~1.y
'" or
DBlO Automat.d
N(!
PrOll!. O.t./T~ To D.t./T~ Start O.t ./T~ D•• tin.tion Name
Li . t Sub.cription Name Li. t
N)
'"'---
T~out
2010- 12-20
+
201 0-12-23 M-RP201 0
D251130_
02 : 00:00
D251131_
D251132_
c-nd - > Ent.r-PP1- - -Pr2---PP3---PPf---PP5---PP6---Pr7---PP8---PP9---PPI0--PP11--PFl2--....it
Figure 141:
ADARPL Replay SOC4 - RPL Cleanup
A valid TOKEN was given when only a few files were selected to be replayed.
7
Monitors
7.1
What can be monitored?
Assume that you replicate from Adabas on the mainframe to Adabas on Windows, you can monitor the following parts:
eireuit
WCP ,.3.3
...
8.1.3
main fra me Replicator on the
mainframe
3
6.1.8.03
Entire Net-WorK or WCP on Ihe main-
Network I line or circuit
SYSAOS { Apas/lnsight I T RIM I Review
8.1.3
$Y$AOS { Apasllnsighl l
3.2.1
TLOG transaciion
6.2.1
Trace Hour1y Snapshot
frame 4
TARGET OB
Monitoring Re plication Parts - Graphie
Adabas on the 2
!
-,...-'-. ~ ~ BU,.,.>
Monitoring - Number of Commands per Second
Ln this case, the following measurements will improve the throughput:
• lncrease the Adabas buffer pool (LBP) on both subscription database and Rep!icator Engine. LBP on Reptor should not be defined smaller than on the subscription database • lncrease the replication buffer (LRPL) on both subscription database and Replicator Engine. LRPL on Reptor should not be defined smaller than on the subscription database • lncrease the threshold of the replication buffer (LOGINPUITRANS) on Reptor from 50% to 70%. Adabas commands can be longer replicated out of the RPL buffer and the usage of the overflow area SLOG can be delayed or avoided. • The number of updates and files per Adabas transaction from multiple users (online and batch) cannot be handled in time from one Reptor. Heavily used files should be dedicated to special Replicator Engines. One company is using for example 8 Replicator Engines. • Activate on the Replicator Engine asynchronous Vol-Ser I/Os (ASYTVS=YFS) and decide to set the parallel LFIOP I/Ds (FMXIO) to a value greater than 1 (normally 6-8) if data to be written and read from the SLOG file.
7.1 What can be monitored?
7.1.1
121
Source OB on the Mainframe
SYSAOS shows the high water marks (HWM) of some Adabas queues and buffers. A Session monitoring / U display resource utilization / H high water marks Critically important for the replication process is using the attached buffers (NAß) and the replication buffer (LRPL). An overflow of these buffers on the source or subscription d atabase w ill stop the replication process and an INITIALSTATE or REPLAY is necessary. . . . . . A D A B A S BASIC SERVICES " . . * High Wa t e r Mark .
11 :f3:44 DBID 39
, ,, ,, ,,, 'W, , ,, ,, ". (LRDP)l ,
Pool/Queue
Attached Buft:er(NAa) Coaa.and Que ue (NC) Format Pool (LI"P) ,~, Hold Queue XSN-Liat Table (LX) 'oq. ~. Tab1e (LQ) U. e r Queue Unique DE Pool (DUQ) Sec:u rity P ool (LCP) (UQF) UQ File Liet ATM Tran e. (:UD) (LWP) Work Pool a.do Pool ,~) Replication
Figure 145:
,
Si z e 2.5760 16800 2500000 560056 4000 9000 466312 60000 10000 1 44360
I I I I
,,, , ,, ,,
I \ Ueed I
uno
I 960 I
"
U28 I
" ."
6160 X
""
1 I
576 I
1500000 I
59908 I
314512800 1
131375292 1
20.3 1.2 o 5 o 7 3
,, ,, ,,, ,, "" ,"
2010-09-2. PACUB02
Date x I x I X I X
2010 - 09 - 2. 20 1 0 - 09 - 2. 2010-09-2. 2010-09-24 2010-09-24 2010-09-24
T ......
11 : 29 : U 11 : 19:59 l1 : 31 : U 11 : 19:59 11 :35: 14 11 :30:54
3 I 20 1 0-09-2. 11 :30:5 4
" " " "
9 I 2010-09-2. 11 : 29 : U 0.0 X
I
, I X I X I
,,, ,, ,, I
..
..
Monitoring - High Water Marks -Source OB
SYSAOS with the HWM funchon also runs in batch and a Natural parse program analyzes programmatically the data (see chapter 7.2).
7.1.2
Replicator on the Mainframe
SYSAOS
The Replicator Engine or Reptor contains of the Adabas nucleus and some Replicator modules. SYSAOS can also be used to show the high water marks (HWM) of some Adabas queues and buffers. A Session monitoring / U display resource utilization / H high water marks Critically for the replication process is using the attached buffers (NAß) and the replication buffer (RPL, parameter LRPL). An overflow of the attached buffers (NAß) w ill stop the replication process and an IN ITIAL STATE or REPLAY is necessary. A replication buffer (RPL) overflow w ill deactivate the destination if no SLOG is defined or activated. The parameter 'LOGINPUITRANSACfION' can be set to
7 Monitors
122
use its SLOG system file as temporary storage loeation for incoming compressed replication transactions, beiore they are queued for processing. A value set to 70%, for example, will automatically switch to the SLOG if the threshold is reached. Without setting a threshold, the SLOG won't be used, even if the SLOG is created and available. The Adabas buffer pool (LBP) is also important but its usage cannot be displayed with SYSAOS on the mainframe, in contrast to Adahas on Windows, see Figure 199. 11:14 : 45 DBID 302
••••• A D A B A S
Atu"h .. d But'!'.. " (MAll)
C_nd OUau.
(Ne) Fo ....... t 1'001 (LW) (NH) Hold Queue 15M- Li . " Tabl. (LI) Ood . Tabl. (LQ)
'~ .
U•• " Qu.u.
(~)
Uniqu.. DE Pool S . curity Pool UQ ril. Li . t A~ Tran • . Work Pool _ Pool
".
a.plic:-tion
Figure 146:
BASJ:C
SERVJ:CES •••••
HiG'T..
."".. ,. .." .."
"
,-_.
~-=:maJ:a~
I..
..
1-·
I::.:::;' ...., .~ .. ...... ~'" ""· "'''1
""""" .
~ ~
*..,•." -*. " ,tl
~ ~
......
' 0.0'
,ut
.~.
-~,
-" ",... 111.0'
.U~
-
, ~.
",21
"' " ,n
'"
I :: :::.~'
,-, ..
~ mli~ll i m
Monitoring Firewall Analyzer - ManageEngine
-
..... ,.... ....... '... ... · ·,.. 1
~
~
53 ; 25 53:26.005 53 : 21.005 53,28.005 53:29.005 53 : 30.005
""" "" ""
""" " "
'" '"
ba1660 ba1560 ba1560 ba1660 (b5abO abs.bO 'b58bO (b59bO ba1660 ba1560 ba1560 ba1660 ba1560 ba1560 ab58bO 'b59bO 'bSabO ba1560 'b5abO ~~
ba1560 'b58bO
,n, ,U, ,n, ,n, ,U, ,n, ,n, ,U, =. ,n, ,n, ,U, ,n, ,n, ,U, ,n, ,U, ,n, ,U, ,u, ,n, ,U, ~. ~. ~.
~. ~. ~.
~. ~. ~. ~. ~.
~. ~. ~. ~.
,~
~. ~. ~. ~.
~. ~.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
N... ADABAS N... ADABAS
_u _u
_ _ N. .. ~ cdl _ N... ADABAS _ N. .. ADABAS _ N. .. AlIA8Io$ _ N... ADABAS _ N. .. ADABAS _ N. .. ADABAS _ N... ADABAS _ _ N. .. ~ _ N. .. ~ N... ADABAS _ N. .. ADABAS _ N... ADABAS _ N... ADABAS _ N. .. ~ c.11 _ N... ADABAS _ N... ADABAS _ N. .. AlIA8Io$ _ N... ADABAS _ N... ADABAS _
_u _u _u _u _u _u __u u _u __u u _u _u
_u _u _u __u u
ADABAS ADABAS AlIA8Io$ ADABAS ADABAS AlIA8Io$ ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS ADABAS AlIA8Io$ ADABAS ADABAS
10-09-13
1619
Natural Parse Program for
N... N. .. N. .. N... N. .. N. .. N... N. .. N. .. N... N. .. N. .. N... N. .. N... N... N. .. N... N... N. .. N... N...
wer on Windows Log
11:23:26
"
_u _u _u _u _u _u _u _u _u _u .,.11
_
.,.11 _ .,.11 _
.,.11 _ "",-11 _
.,.11 _
.,.11 _
_u _u
""'-11 _ .,.11 _
.,.11 _ 23:26
7.2 Self-Written
7.1.7
129
Adabas on Windows - Command Log
As explained in chapter 7.1, it is very helpful to determine the number of Adabas commands per second on the target database to analyze bottlenecks. ADAr18 REC
79219 79220 79221 79222 79223 7922' 79225 79226 79227 79228 79229 79230 79231 79232 79233 79234 79235
N2 Ort by " a inq SYSAOS I R40plication Maint.nanc. ---* 11*------------------------------ -----------------------------------. IIRPLSTATI EXBC i'QI-HATBATCR, RBQION-OM, P_'P_SlPROO,SYS-STRSYS,AIJ'l'O-ON,lM-O,STACJtoo(LOGON SYSAOS ; MENUj' IISTBPLIB 00 OISP-SHR,D~ . ALIAS.SIPROtl.LOADL:IB ADABAS 11 OOOISI'-SHR , OSNooAOABAS . ALIAS.SIPROtl.PGKLIB NATURAL 11 00 OISP-SHR,D~ . ALIAS.SIPROtl.PARMLIB NATPARM IloDCARO 00 OISI'-SHR , O~ . CONTROL.CAPl)(PRODN) IICMPRINT 00 OSN-DIETBR.SLOGREP3, 11 OISI'- (, CATLG) , SPACE_ (TRK , (l0, 10) ,RLSB) IIc:wsmIN 0,302
11
,,
"
.
B , ,rIN
/.
11* -------------------------------------------------------------11* Ch.ck SLOG ro" .-pt.y (RC-O) 0., not. .-pty (Rc-12) 11* -------------------------------------------------------------IISLOGRBPI BXEC NATBT,OEPT-ST,OB-PRST,Srsour-*,TIME_IOO, 11 _ ' IM-O ,MlUIID-O ,MAXCL-O ,MT-O ,LT-~99~99 ,AIJ'l'O-ON, Ou-ABEND ' IISYSOUT
00 SYSOUT-* DD DISP-SHJl,OSN-DIET&Jl .SLOGREP3 IICMPRTOI 000SN-DIETBR.MSG1, 11 D:IS P- (,CATLG) ,SPACE- (TRK, (10,10) ,RLSE) IISYSIN 00 * LOGON PDS'l'UTIL OWSLOGCK 11~1
... 11 11
BNOll' Ir SLOGREPl. N Re _
° TREN
11** ---------------------------------------------------------------11** E-mail N. two .. k . ndOB . . . . .vailabl. 11** ---------------------------------------------------------------IIEIQ,IL2 EXEC PGM-llUllPTOl ,RBGION_40~61t, Onu.MllBR_SO IISYSTSIN 00 *
136
7 Monitors
SMTPNOTE TO(O$'1'
[email protected] , e . ALIA$.$IPROD.LOJIOLIB DD OIS_SHR,D~ABAS.ALIAS.SIPROD.PGKLIB DD DIS_SHR,D~ . ALIA$.$IPROD.PARMLIB OD OIS_SHR,D~ . CON'1'ROL.CAJUl(PRODIl) DD DSIl-DIETER.$LOGREP4, OIS_ (, CATLG) ,SPACK- (TM, (10,10) ,RLSB) DD *
SYSAOS;MImU)' ADABAS NA'l'URAL
RATPARM
0,302
, ,TIN
/.
11* ------------------------------------------------------------------11* Ch.ck SLOG for ..opty (RC-O) or not ..opty (Re_12) 11* S ...... SLOG report 1 and SLOG report 2 empty (Rc-O) SLOG empty 11* S"'" SLOG report 1 _ ' , .nd rep 2 EQ (Rc-O) ..opti-.l. 11* S"", SLOG report 1 GT ....... SLOG r.port 2 (Rc-O) empty in proq 11* S"'" SLOG r.port 1 EQ ....... SLOG raport 2 ( Rc-12) open <Mo.tin 11* open did not work 11* ----------------------------- ----------- ------------ --------------IISLCGREP2 EXEC NATST,DEPT-ST,DB-DEVT,SYSO!J'r- · ,TlME_lOO,
II
nc-o ,MADIo-O ,M1oXCL-O , _ 0 , LT-999999 ,AU'l'o-oN ,DU_ND ' $YSOUT-· DIS_SRR,OSllaOIBTER SLOGREP3 DISP-SHR,DSN-DIETI!R SLQGRI:P4 OSN-OIB'l'ER.MSG8, 01SP- (,CATLG) ,SPACE- ('l'R:K, (10,10) ,RLSE) OD * PRM- '
IISYSCNT //CMIftI.POl
IICMIftI.P02
//CMPRTOI
II
IlsYSIN
DD OD DD 00
OWS:LOGC2 ".
/I /I
Ir SLOGREP2 !l.Re _ 0 THEN
IIEMPTY
BXBC PGM-IKJZPTOl,REGION_4096X,O~50
SLOG .... -..pty
11* · ---------------------------------------------------------------11** EMAIL SLOG EMP!'I 11* · ---------------------------------------------------------------IISYSTSIN
OD .
SlftPNOTll TO (
[email protected]) SUBJECT(Replicator: SLOG D8302 i . _ t y )
BATCR OJ.TASBT ('OIBTI!R.MSGS' ) SlftPNO'l'l: TO(LrRAZIN@CALSTRS , COM) SUB.JBCT(Replic.tor: SLOG 08302 i • ..opty) BATCR DATASET ('DIBTEit,MSG8') SlftPNO'l'l: TO (
[email protected]) SUBJECT(Replicator: SLOG D8302 i . empty) BA'l'CH OJ.TASBT ('OIBTI!R.MSGS') IISYSPitOC DD DSN_ $IS2.CLIST,D1SP-SKR IISYSTSPRT 00 SISouor-* 11 ENDI" /I Ir SLOGREP2 . N Re _ 12 THEN
11*· ---------------------------- ----------- ------------------------I I **
EMAIL SLOG NOT BMPTY
11* · - - -------------------------- ----------- ------------------------IINO'l'BMP'l'I BXBC PGM-IKJZrrOl, REGION-4096X,DYN1oMNBita50 IISYSTSIN
DD.
7 Monitors
138
SM'l'PNO'l'II: TO(DSTORR@STORRCONSI1LTING . COM)
SU8JZCT(Roopl DB312 , SLOG 1. not. ampt.y DAT~T(
-
Open da. tin. raU..:!) BATe" -
'OJ:BTER . MSG8')
SMTPNOTI! TC) (
[email protected]) s=cr(Repl OB312: SLOG i . not _ t l ' - Open eIe.tin . f:-.iled) BATeH
OATASIIT ('DIETIIR.MSGB·) SM'l'PNO'l'II: TO (ADABASSERVERADKIN:lSTRATQilS @CALSTRS . COM) SUBJECT(Rapl DB312 : SLOG ia not alpty - Open EXEC NATBT ,tlEPT-ST ,tls-tJEVT , SYSOOT-', TINS_100 , _ ' r-D ,MIUIIo-O ,Ml\XCL-O ,Ml'-O ,LT-9999gg ,AlJ'l'O-ON, DU-ABEHD' IICMNIU'Ol tltl DISP-SHR,tlSN-tJIETER.RBPF30.tlB312 input 11C'IIIKU02 00 DSN-DIETER.RPRJO.I1BJ12, output ADS report 11 RECrM-FB ,LRJ:CLoo79 ,BLltS IU_O, UNIT-MlSCDA , 11 DISP-(,CATLG) ,SPACE-(TRIt,(10,10) ,RLSE) 11C'IIIKU03 tltl DSN-oIBTER. kINFJO. tlB312 , output ot RPLJOREP 11 RECPMooFB,LRJ:CL-080,BLKSIZE_0,UNIT-MISCDA, 11 DISP-(,CATLG) ,SPACE_(TRIt,( 10 , 10) ,RLSB) IICMPRlNT 00 SISOUT-* IICMSYNIN 110' LOQON STIlBJWTI RPL30REP
11
". 11
I1IBTBR.RJ:pr30.DB312 output of 8ISAOS report tlIETER.RPFP30.DB312 "el.vant linea, input tor RPL30REP I1IJ:TBR.RINr30.I1B312 output of proqr_ RPL30REP
Figure 168:
SLOG Number of Records and Space Check OCL)
7.2 Self-W ritten
141
0IETER.RlNr30.0B312 •• • • •• •• •• ••• • • ••• • •• •• •• •• •• • •• • Top ot Oat. • •• • ••• •• •• •• •• •• • •• •• • ••• •• • • ••• • 2010-12-16 - 17:55:03.7 - Raport ot SLOG Vi1. 30 OB312 - RPL30REP Ra"ord. 10adad 0 Ul . pao* ,RI!:cns-vB, LRBCI.-2tO 000017 IIEJZSl:N DD· 000018 ST REP'noIPH 000019 rum RETMi'M 000020 :S 000021 rl:ND DOPRINT 000022 : 1: 000023 11
Figure 170:
Read Output Queue of Active Job - EjESLNK QCL)
See the parse program at the appendix, chapter 9.5.2.
7 Monitors
142
The report of the parse program: 017528
-.........
,,. '" -" '" -" '" ,,. '" -" '" -" '" -" '" '" -" '" '" '" -" ••••••••••••••••••••••••••••••••••••••••••••••••••••• '" J:T cmd to DSID Al cmd to DBID J:T cmd to DSID
017529 017530 011531
tT cmd to DSID
017532
Al cmd to DBID IU cmd to DSID B1 cmd to DBID
011533
017534 017534
A6
~
~,
~
18 RSP
~
~, ~,
18 RSP 14 RSP 14 RSP
-POOL
.W>oodo . W>oodo .W>oodo .W>oodo • W>oodo .~
.~
, ,,• u. , ".
.
". "
m
'"
237158
237186
••••••••••••••••••••••••••••••••••••••••••••••••••••• ••••••••••••••••••••••••••••••••••••••••••••••••••••• 011534 RPL - POOL LRPL- 157286400 149875480 ( 95 ' ) ••••••••••••••••••••••••••••••••••••••••••••••••••••• PNR/RSP/CNT: PNR/R$i'/CN'l':
PNR/RSP/CNT: PNR!RSP/CNT:
PNR/RSP/CNT: PNR!RSP/CNT:
PNR/RSP/CNT:
Figure 171:
, "" '" '" "" '" n '" "" '" "• 254 4
113 4
254 4
Output of Natural Parse Program
EJESLNK and the parse program can be used to check DDPRINT of the source database and the Replicator Engine.
7.2.7
Target Check - Ping
The following job is scheduled to run every hour to check whether the target database, Adabas on Windows, is reachable and the files are available. A Natural program reads one record of one file on the target database and set based on the Adabas response codes a condition code for the batch job. See Natural program in appendix chapter 9.5.1, ring Target Database.
7.2 Self-Written
IlpINQ JOB I I *!GIN
143
(ST345T, ST01, ST) ,DBA,MSGCLASS-l, TIMI!:_NOLIMJ:T Now-NO, SISTEM-SI5, CLASS-plO ,USl!:R-DIB'l'ZR PR,DONAMZa,FORM.Sa41, 1'CBaA8 ,CIIARS-G125
I/*!'ORNAT 11** --------------------------------------------------------------- **
I/u SCHEDULEtI I'OR OAILY EVl!RY HOUR -- DI!;TI!R.PJLSRVCIU!Ck (TEMP 0"')"
11**
I/u CHZCX WlU!TIU!R TAlUZT OB IS UP
AN!)
.*
RUNNING - SEND EMAIL
Ilu
I/u ,An
Ilu I/u 2010/12/06 /lH 2010/11109
I/u 2010/11/02
Ilu 2009/12/01 /lu
OSTORR DSTORR OSTORR DSTORR
PREPAREl) I'OR SY5 II:l.IMlNATED SRP EMAILl WITN ROooO CIU!CIt DATJlBASB AN!) MESSAGE DATAS1!T, EI' l!!VEN'l' CRKA'l'BD
11** -------------------------------------------------------------IIPJLCHECIt I!!XI!!C NA'l'BT , DEPT-ST ,Da-DEVL, PRM- ' IM-!) ,AUTOooON' IISYSOUT DD SISOUT-*
IICMPRINT II /I
OD DISP-(,CATLG) ,DSNaDIE'l'ZR.DB251CH!t, UCnc-FB, LRl:CL-120 ,BLll.SIZE_ O, UNIT-KISCOA, SPACB_('l'PJt, (10 ,50) ,RLSB) IICMSININ DD * LOGON DIETBR DWSRDUR 251 ".
1/** ------------------- - --------- - ---
II
11' ROooO TIU!N
1/*** --------------------------------Ill!:HII.ILl I!!XI!!C PGHooIlW1:rr01,RBGIotl_40961t,D~50 I/SYSTSIN
OD *
SM'l'PNO'l'II: TO(
[email protected])
SUBJEC'l'(PJL: OESTINATION DATAAIlSE 251 DATASET (' DIB'l'J!P.. DB251CHll') I/SYSPROC OD DSN_SIS2.CLIST,DISP-SHR IISTSTSPRT DD SISOUT-* I/ ENOII'
-
AN!)
NETWORlt IS UP)
BATeH -
11** --------------------------------11 11' RC GT 11 THEN 11** --------------------------------III!:HII.IL2 I!!XI!!C PGHooIIUIl"01,RBGlotl.4096K,0~50 IlsISTSIN
DD * SM'l'PNO'1'I!! TO(DSTORR@STORRCONSULTING COM)
DBSTINATION DATABASB 251 Oll. NETWOJUI< IS DOWN) BAT(:H DATAS1!T ('DIBTBR.DB251CHIt ' ) IISISP!I.OC DD DSN_ SIS2.CLIST,DISP-SHR IlsYSTSPRT DD SISOUT-* 11 ENDII' s=cr(RPL:
11* * ---------------------------------
I/DELI!TI!! EUC PGHoolEFBR14 /lDDl DD DISP-(OLD , DELI!TI!!) , DSN-DIETBR DB251CH!t
/I
Figure 172:
WCP, Network and Target OB Check - PING - OCl)
..•••• ..•• •• ..••••
144
7.2.8
7 Monitors
Destination Report
The following job submits SYSAOS in batch and creates areport, which will be parsed by a Natural program. 000001 I/STDBSl02 JOB (STl4ST,STOl,ST) ,DBA,HSGCLASS_l, 000002 11 000003 //*HI'oIN 000004 I'*I'ORNAT
NOT l n_,SISUID, TINZooNOLDaT , TlPRUN_SCJIN KOLD-NO,SrSTDPSl7 ,~S.i'10,USBR-DIBTER PR,DD~.I'ORMS •• l,1'CBaA8,CHARS-C125
1/*-----------------------------------------------------------------0
000005 000006 // * a.p11cator Subacript10n Check - DIBTER.CNTL(DBST302 ) 000007 //* ADA9 1 3
//*-----------------------------------------------------------------* //*-----------------------------------------------------------------* /IIlBLBT:E EXEC PGMoo I DCAMS
000008 000009 //* !)elat. dat ••• t. t.,0111 prior r»n 0000 1 0 000011 0000 1 2 000013 000014 000015 0000 1 6 0000 1 7 000018 0000 1 9 000020 000021 000022 000023 000024 000025 000026 000027 000028 000029 000030 000031 000032 000033 000034 000035 000036 000037 000038 000039 000040 000041 000042 000043 000044 000045 000046
//SISPRINT
DD SfSOUT-* DD * DELBTZ DIETZP. . DEST302
Ils!Sl:N
PURGIl
--------------* --------------*
11*--------------------------------------------------11* Cr.at. SUBS report by uaing SfSAOS 1 Rap1ication Maint.nanc. ---. 11*--------------------------------------------------IIRPLSTATl EXEC PGMooNATBATCH ,R&GION_OM,
11
PARM-'PARM-SfPROD,SfS _ STR$f7,~N,~,ST~(LOQON
IISTIIPLIB
11 11 II DDCARD IICHPR I NT 11 IICMSYNIN 0,302 X ,. D
DD DD DD DD DD
DISP-SHR,DSN-ADABAS.ALIl\.S. SIPROD. LOADLIB Dl:$P-SHR,DSN-JWABA$ . ALIA.$.SIPROD.PQa.IB DISP-SHR,DSN-ADABAS.ALIl\.S.SfPROD.PARMLIB DISP-SHR,DSN-JWABA$ . CON'l'ROL . CARD(TESTf) DSN-oIETZP..DEST302, DISP- (, Cl\.T!.G) ,SPACK- (TlUt, (10,10) ,RLSE)
Sf$AOS;MENU)' ADABAS NA'rURAL NATPAPll!
DD
,, "IN
11
Figure 173:
SYSAOS Batch - Display Reptor Statistics - Destinations aCL)
7.2 Self-Written
15:35:20 Raplicator o..tN.....
--------
D251018 D251035 D251035 D251035 D251035 D251144 D251158 D251167 D251168 D2512451 D251227 D251228 D2512251 D251230 D251231 D251232 D2512451
Figure 174:
_._._. _._._. . _._._ _. -. -. '"
"..
145
••••• A D A B A S
2010-10-14 PRPTS04
SERVICES . . . . .
Total raplicatad tran.action.
,
.b. .
Total pandinq trana
'"
2,098
""
3,0511 5,764 5,954 4,000
,, ,,
2,000 2,000 2,000 2,000 2,000 2,000
,
.b. .
SYSAOS Batch - Display Reptor Statistics - Destinations - Output
U:53:06 Rap1icator 302
..... A D A B A S
BASIC
SERVICES' . . . .
- Sa 1a"t.d o.U:ination Stati. ti" . -
D251035 Sa1act.d Da.t Type AcIaba. Total rap1icatad tran.a"tion. Total pandin'1 tran.action. Total me •• aga. .ant Total comm.it • . , •.. Sloq loqgad count Sloq deloqqed count Itema tor de.tination on .1og . :Itema to delata tram .1og Byt.• •a nt to dea tin.tion Pandinq byta. tor de. tination Pa nding me •• aga. Numl:>ar ot BT ' . . ........ . . NUlllbar ot BT ' .
Figure 175:
BASIC
- Da.tination Statiatic. -
2010-10-U PRPTS04
T~ tran comm.ittad 1,062 2010/10/14 14 : 51:16 1,592
,
,,, ,, ,,,
o T~ Of la . t ET 1,062 2010/10/14 15:03:251
SYSAOS Batch - Select one Destination - Output
7 Monitors
146
7.2.9
Subscription Report
The following job submits SYSAOS in batch and creates areport, which will be parsed by a Natural program. 000001 I/STSUBl02 JOB (STl4ST,STOl,ST) ,DBA,HSGCLASS_l, 000002 11 000003 //*HI'oIN 000004 I'*I'ORNAT
NOTln_,SISUID, TINZooNOLDaT , TlPRUN_SCJIN KOLD-NO,SrSTDPSl7 ,~S.i'10,USI!:R-STADBA PR,DD~.I'ORMS •• l,1'CBaA8,CHARS-C125
1/*-----------------------------------------------------------------0
000005 000006 //* bp11c:ator Sub.""ipt1on Check - DIETER.CNTL(SUBS302) 000007 //* ADA9 1 3
//*-----------------------------------------------------------------* //*-----------------------------------------------------------------* /IIlELBT:E EXEC PGMoo I DCAMS
000008 000009 //* !)elat. dat ••• t. t"0111 prior r»n 0000 1 0 000011 0000 1 2 //SISPRINT
DD SfSOUT-* DD * DELBTZ DIETZP. . S08S302
000013 Ils!Sl:N 000014
PURGIl
--------------* --------------*
000015 11*--------------------------------------------------0000 16 11* Cr.ate S08S report by uaing SfSAOS 1 Rap1ication Maintenanc. ---. 0000 17 11*--------------------------------------------------000018 IIRPLSTATl EXEC PGMooNATBATCH , R&GION_OM ,
0000 19 11
PARM-'PARM-SfPROD ,SfS_ STR$f7,~N,~,ST~(LOQON
000020 IISTIIPLIB
000021 11 000022 11 000023 II DDCARD
000024 IICMPRI NT
000025 11 000026 000027 000028 000029 000030 000031 000032 000033 000034 000035 000036 000037 000038 000039 000040 000041 000042 000043 000044 000045
IICMSYNIN 0,302 X ,.
DD DD DD DD DD
OISP-SHR,DSN-ADABAS.ALIl\.S. SIPROD. LOADLIB Ol:$P-SHR,DSN-JWABA$ . ALIA.$.SIPROD.PQa.IB OISP-SHR,DSN-ADABAS.ALIl\.S.SfPROD.PARMLIB DISP-SHR,DSN-JWABA$ . CON'l'ROL . CARD(TESTf) OSN-tlIETZP.. S08S302, DISP- (, Cl\.T!.G) ,SPACK- (TlUt, (10,10) ,RLSE)
Sf$AOS;KENU)' l\DABAS NA'rURAL NATPAPll!
DD
Po
,, "IN
000046 11
Figure 176:
SYSAOS Batch - Display Reptor Statistics- Subscriptions QCL)
7.2 Self-Written
1!i8bO 'b59bO 4bSabO ba1560 'b5abO 'bSabO ba1560 'b58bO
~. ~.
~. ~. ~.
~. ~. ~.
~. ~. ~. ~. ~. ~. ~. ~.
.
~. ~. ~. ~
~. ~.
Analyze WCP Log
,n, -,n, ,n, -,n, ,n, -,n, ,n, -,n, ,n, -,n, ,n, -,n, ,n, ,n, ,n, -,n, ,n, -,n, ,n, ,n, ,n, ,n, -
......... ...... ...... ...... ............ ......... ......... '" '"
10-09-13
~
~
--_u _u _u _u _u -_u _u _u _u _u _u
_U
~
_u _u
~
_U
~
~ ~
~ ~ ~
~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~
~
_U _U _U _U _U
_U
_ U
...... ...... ...... ...... ...... ......... ......... '" '" '" '" '"
'"
~
~ ~ ~
~ ~ ~
~ ~ ~
~ ~
n,23,U
"all
".11 .,.11 "all ".11 "all "all _U "all "all _U "all
~
".11 _U
~
"all
~
~ ~ ~ ~ ~ ~
~
".11 ".11
"all
".11 ".11
"all
".11
----------------
7 Monitors
152
Using a self-written Natural program, you can parse through the wer log and count the Adabas calls per second. YOll can also use a text editor for Windows with search capabilities, such as display number of lines with ' New ADABAS call New ADABAS call ='. A selection function by columns (from/ta) is beneficiary to search a command log size of 200 MB and more.
7.2.12 De1ermine Highesl ISN Currenl ly Used The program ISNHIGH reads fiJes of a database and reports the highest ISN currently used. The Natural program is described at chapter 4.3, Compare Files of Source and Target OB, and is located at appendix, chapter 9.5.10.
7.2.13 Compare File Componenls After Replicalion The following job creates an Adabas report on the mainframe with NOFDT and NOSTD and FTP it to a Windows server. The Adabas on Windows report is created and saved directly to the Windows server to be compared. IIsTREP039 JOB .
11 MSGCLASS_2 ,NOTlrr_'SYSUIO, TIMII_1UO 11* . 11* -------------------------------------------------------------11* OELETE 1 11* -------------------------------------------------------------IIDEUTEl !:XI!C PGIoPIEI'BR14
/1001
/I /I
OD
OSN-ST.AOAOBA.AOAREP39, OISP-(HOO,DELETE), SPACE-(TiUt, (1,1) ,RLSE)
11* -------------------------------------------------------------/1*
ADABAS REPORT VB - DIETER.CNTL(AOARI'039)
11* ------ ----------- ----------------------- ---------------------IIRZP039
BXZC AOAUSRM,0S-STST9 , UTILITY~P , QUAL-MPM, VBR-v81 II00TBHPRl OD omea II00sORTRl 00 omea /lSTl . DOPltINT 00 omea IIST1 . ODOR!}(:!\. DO OIS P- (,CATLG) ,SPACE- (TJt:K, (1000,10) ,RLSE) , 11 OSN-DIBTER.AOAREP39 IISYSIN 00 * ADAP.&P RZPORY AOAREP NOftlT, NOSTD
11* -------------------------------------------------------------/1 * rrp TO THE Z-ORIVlI:
11* -------------------------------------------------------------/lnPl
!:XI!C PCMooI'TP,RZGI0Na4N,PARM-' (TINEOUT 100 TRACB EXIT' 00 SYSOUT-* IISYSPRINT 00 SYSOUr-* IISYSTCPD 00 OIS_SHR,OSN_ SYS1. TCPIP . PARNLIS (TCPDATA) IlsysrrPD 00 OIS_SHR,0SN-SYS1. TCPIP . PARNLIS (crrPDATA) IlrrPINl 00 OIS_SHR,OSN-oIETER.AOARI!.P39 IIINPUT 00 OISP-SHR,OSN-OIETBR.CCAROS (I'TPLOOON) /I 00 *
II000PUT
.scii
~~,
pgt /loo:rrPINl
ADAP.&P39_FILE_RPL . 20110127 txt
~" "0" 11
Figure 183:
Create and FTPanAdabas Reporttoa Windows Server
7.3 Optimize for Infrastructure (041)
153
The program RPL-P002 reads both Adabas reports and creates a list with TOPISN, Records Loaded, and MAXISN of both source (DB039) and target (DB251) databases. See the Natural program at appendix, chapter 9.5.13. FILa-NR
000006 000007 000008 000009 000010 000011 000012 000013 000014 000015 000016
TOPISN 039 TOPISN 251 RBCLOD 039 RBCLOD 251
~ISN
039
~ISN
251
---------- ---------- ---------- ---------- ------------------1 ,271 1,279
'"
.
195,637 1 ,098,767
,"
'"
195,637 1,098 ,767
"."
25,665,942 25,665,942 250,289 250,289
,
6,004,159 2,439,377 3,887
Figure 184:
,
6,004,159 2,439,377 3,887
'" ,"
.
195,635 1,098,767
'"
.
95,635 1,098,767
1,000,427 196 ,095 1,251,011 1,114,879 1,271 2,047 3,815 25,191,478 25,191,478 30,000,119 25,673,727 249,229 249,229 1,000,427 257,279 8,191 6,003,833 6,003,833 8,000,243 6,004,735 2,439,314 2,439,314 6,000,023 2,4 41,215 3,673 3,672 5,087 8,191
,
,"
,
'"
'"
Create Adabas Report and FTPtoa Windows Server
The report shows differences in highest ISN (TOPISN), records loaded (RECLOD) and the maximum set ISNs (MAXISN). One record of fil e 16 received RSP 113 and was therefore not replicated, as pointed out in Reptor's log dataset DDPRINT.
7.3
Optimize tor Intrastructure (041)
7.3.1
General
Currently, companies have to write or purchase several monitor programs to gather statistical data to evaluate the performance of Software AG products. They use this data to improve their processes. To make this more effident, Software AG and other vendors offer single monitor software on the mainframe such as Review, Adabas Statistic Fadlity, Adabas Online System, Adabas Event Replicator Subsystem, Apas/Insight, System Management Hub, and Trim. Some customers find it very awkward to get performance data f,om a green screen on the mainframe and then switch over to the PC for additional information on Windows-based Software AG products. For example, if you monitor Event Replication for Adabas on the mainframe, you must dose the Adabas Event Replicator Subsystem screen and open the Adabas Online System screen to get additional information for the same product.
154
7 Monitors
Ln 2009 Software AG introduced their ' new tool' Optimize for Infrastructure: the Mainframe Edition. For many years, webMethods u sed their Business Activity Monitoring (BAM) tool, which includes Optimize for Process, Optimize for SAP, and Optimize for 828. After the merge, Software AG updated and improved the existing Optimize product for monitoring their enterprise products in real time. This tool is UNIX or Windows-hased and collects data from Software AG's produets such as Adabas, EntireX, Event Replicator for Adabas, Adahas Delta Save, Natural, Entire Net-Work and A dabas Fastpath. The benefit of this solution enables yOll to observe yOUf performance data from a single weh-hased dashboard using YOUf favorite Web browser.
The Software AG products Adabas, Event Replicator for Adabas, Com-plete, Entire Net-Work, Natural, Entire System Server, ApplinX and EntireX are already equipped with ' hooks' to get performance data if the latest versions are installed. Ln some cases, small system modifications (ZAPs) are necessary. One or more special Natural RPC servers on the mainframe are necessary to get the data horn the above-mentioned products through EntireX Broker (webMethods server) to a JMS Queue and on to the Analytic Engine for interpretation and application for KPI rules. The Analytic Engine stores the data into a relational database. For example, a batch Natural RPC server monitors all global Natural components: a CICS Natural RPC server monitors all global Natural components: a CICS Natu ral RPC server monitors in general only local environment-specific Natural components. If EntireX Broker is not available on one of the rnonitored platforms, the EntireX Broker stubs (EntireX Mini Runtirne) must be installed.
Figure 185:
Analyze wer Log (Sourre: SAG)
7.3 Optimize for Infrastructure (041)
155
Optimize for lnfrastructure comes with over 600 predefined Key Performance Indicators (KPIs), which can be selected and activated to collect the data you want to monitor and analyze. You can also create rules, send predefined alerts or start activities by using Web services. ln addition to the data collection, Optimize implementation performs other main functions such as data communication, data processing and analysis, data storage and data presentation . Optimize for lnfrastructure uses the following parameters to monitor Adabas Event Replicator: Dimensions, KPls and Built-in Rules.
7.3.2
Dimensions
AdminIstratIon Name
Olsplayed Value
Host
The name of the host.
Product
The product component name "Adabas Event Replicator".
AdabasEventRep
The 10 and name of the Event Replicator server.
7.3.3
KPls 01 Event Map AdabasEventRep
AdminIstratIon Name
KPI Name
Umt
Type
DescnptlOn
RepBytesSentTotal
Bytes Sent Total
count
delta
The total number of bytes sent in the messages.
RepFreeSl OGAssoSpace SlOG - ASSO Free Blocks
percent last vatue
Free space on ASSO for SlOG entries.
RepFreeSl OGDataSpace SlOG - OATA Free Blocks
percent last vatue
Free space on OATA for SlOG entries.
RepHWMRepPool
percent last High-Water MariSOO N'J'a90 LPP-IM LWP-7000000 LI_ SOOOO
(RBP'lOR)
REPLICl\.TION POOL rCIR NUCLEUS PJ!:PLlCATION SBT Ta TBS POR 'MIIS DATABASB
cr
CACHE $IZE CHANGE
--> --> -->
PLOG8 JUUI: REQUIRED NO OPEN REQUIRED NOT A UTILITI ONLI SBSSION ~ or CQJ:'S WAS 300 8/23/03 ~ OF TRRBADS WAS 105 08/23/03 L1!NGTH OP PI POOL !'ROM 800000 6/21/03 LKNGTH or W01Ul POOL WAS SM 11/22/08 L1!NGTH OF TBI POOL WAS 30'" 11/03/07 NT-WOU DATABASB Ll!NGTH or TBLI!S POOL WAS 260'" - 6/4/05 L1!NGTH OF SORT ARM lOB LBNGTH (MIU) Jlu.D ONLT NUCl.oZUS ISSUZ CltPT ZVBRT 5 MlN .ACS 9/06/03
TIHJ:OU'1' INTZRVlU. TRANSACTION TIMZOUT INTZRVlU. WAS 600 lNAC'l'XVITT TIHJ:OU'1' IN'l'J:RVAL WAS 900 INAC'l'XVITT T:tMBOUT INTJ:RVAL (BTaT) lNAC'l'XV1TI TINBOUT INTZRVlU. (BXUU) AT'I'ACIIED Burn:RS WAS 250 8/23/03 ---> SBCURITT POOL SIZB ISNS PER TBI BLBMZNT PlUUI.LLI!:L 1/0 OPERATIONS WAS 30 8/22/03
COMMNI1)
--->
NUCLEUS POOLS
DATABASB/USBR L~TS
ADARUN NQCID-80 ADARUN NISNHQ-5000
--->
--->MIU HBLD/USER«
MIU N\JHBBR CIDS/USER 2 5 \ NH)WAS 35 00 6/24/06
D. W. Storr, Efficient Usage of Adabas Replication, DOI 10.1007/978-3-8348-8214-1_9, © Vieweg+Teubner Verlag | Springer Fachmedien Wiesbaden GmbH 2011
9 Appendix
170
--->TIM!!
ADARUN TLSCMD-300
LDCIT I'OR SX
~
• PRO'1'ECTION LOGGING
ADARUN PLOG!IEV- 83ii3 ADARUN NPI.OGoo4
ADARUN PLOGSIZZ-30000 • caeo.ND LOGGING
_w
_W _W _W _W
_w
_W _W _W
LOGGING-YBS
"" ~" ~" "" ~" "" "" ~" ~" "" ~" "" "" ~"
~m
""~W
""_W
~m
"""",""m
LOGXB-OO LOGIo-nS
""
"""","m
• USER UITS
ADARUN UEJ:4-DBALGRIS ADARUN UEJ:S-t1EXIT08
~
=o~
~
-~
~~ ~~
~ ~~ ~~
UB EXTENSION
•
~
•
'/0
•• nn -~ • ,. -~ w -~ • • ,. -~ -~
~,
ADARUN UEl[12_Ul!XI:T12
USER Ul'rS USER ZZ:I'U 2
ADARUN DSFZXl-tlSFZXIT
DBLTA SAVE USER UIT TO SUBlCIT
• DTNAMlC CI' NO DBl.TA SAVE REQUIRI:D 4_> PLOO Rl:QUlRED 4_> NO OPEN Rl:QUlRED 4_ > NOT A UTILITI ONLI SESSION RBAD OIlLI NUCloBUS
ISSUE sms 60 C1tPT 15 MIN .ACS 9/06/0] --> MODIPIED 8Y JNA, PRJ:V-~YBS
--,
• NUCIJruS POOLS
•
ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN ADARUN
• • • •
NC>200 NH_ 20000 NT-18 NU-750 LBP>o100000000 LFIOr-25000000 TFLUSa-O LPP>o600000 Lwr-fOOOOOO LS_ 60000 LI_lOOOO LQ-l00000 LU-200000 NAB-820 LCP>o2000 NSISN-51 LDEUQr-250000
TIM!! LIMITS
OP CQE' S OP HQB' S OP THREADS OP USERS LIINGTH OF BUPFI!:R POOL LIINGTH CP UrNC BUPPE FLuSN POOL BUF FLUSH DURATION - 0 WHIm LPIOP IS USBD LENGTH OP PI POOL LIINGTH OF W01Ut POOL LIINGTH OP SORT AREA LENGTH OF TBI POOL LIINGTH OP TBLI!S POOL 108 LKNGTH (MlUI:) NUMBER OP ATTACHED Bt1!'PERS ---> SECURITl POOL SIZE ---> ISNS PER TBl ELEMENT USED BY SLOGGrNG TO • LIUlZL' MSG COMPONBN'l'S WAS lOOK 08/18/2007 NUMBER NUMBER NUMBER NUMBER
i .... .
.....
s:
Vl
.g
~
~
....
I
...
~
t;
'"
!i
Oe
"
~
8
~ ~
[; i,
H
;~ ~ ~ ~! 8
~~
~i
1a"O
'$;'05;2J.'~2'l'"
Figure 200:
Replication Process - Number of Commands per Second (cps)
178
Figure 201:
9 Appendix
Replication Process - Number of Commands per Second -Spreadsheet 1
9.4 Test Results
Figure 202:
Replication Process - Number of Commands per 5econd - Spreadsheet 2
179
180
9 Appendix
9.5
Natural Monitor Programs
9.5.1
Ping Target Database
0010 DEPlNE DATA LOCAL 0020 1 FJ:Lll:U VIEW 01' $T-RZ-CN'l'R!I-LlNE
0030 0040
2 LOG-COUNTER
0050 , '0. , '0' 0070 , 0060
0080 0090 0100 0110 0120 0130
•
2 ENTITY (Al) (NS) .~ (14)
INIT (' S'> IN:lT
'*
t'or UDB USRI040N
S_ •• t G-gat
'" f'or UDB
JmI)-DEFlNE
••• ------------------INPUT 'DB ••• ------------------ON ERROR
0140 0150 0160 0170 0180 0190 0200
Ir *ERlIDR-NR • 3148 WlUTJ: (0) NOTITloE ' BP.P.OR: Oataba •• ' 'OB ' or Nat-Worlt i . down ' TERNINA'l'E 12 &LSE WR.ITE (0) NOTITLZ 'Otb
•• •• •• •• •• •• ••
-HWM>'5 ' rNR(1:999) - RSP and CNT (1:255)
Array :
.. Adab.. ..*. ---------------------------------------------------------------------•• Date
----------
•• 2009-02-18 •• 2009-04-13
••
DSTORP. OSTOPR
Cop1ad ~rom p.r~ormanc. proj.ct PER-POOl ODPRlNT haB no lon!i.r an l\.SA chara"t.r
OEFINE DATA LOCAL Inp1.lt Work ril. 1 'ADALOG-LINE (11.133) 1 REJ;lErINB 'ADALOG-L1NB • 2 'ADALOG- LINE-ASA (11.001) 2 'ADALOG-L1NE-TBrr (Al33)
••
1 'DUTPUT-L1NB (Al20) 1 REJ;lEFINE 'DUTPUT-LlNE 2 'BEQ (N6)
..-
2'~
(1.2)
2 'TEXT
(All2)
2 'RaP
(N3/1:255) (N6 / 1 :255)
,•• ,, """ tu ,, '" "
"" "" "••
rNR _ OrJl'SET IN TABLB
(N4) /. rNR INDEX / . RaP/CNT (N4) (L) INIT (L)INIT 15 , --------I r SUBSTRING('ADALOG-LINB-~T,01,17) _ 'AB -POOL AND SUBSTRING('ADALOG-LlNE-TEXT,42,02) GT '75' ADO 1 TO 'LC 'ADALOG- LlNE - TEXT _In: WOlUt FrLl: 2 'ASTJ:RISIt WRITE WOlUt PILI: 2 'OUTPUT - LINJ: _In: WOlUt "ILI: 2 'ASTJ:RISIt
"" "" "" "
'Re :.
..
'TEXT :.
..
'HWM :_ TRUl!:
END-I:r Ir
SUBSTRI:NG ('ADALOG-LINJ:-TErr, 01,17) _
'RPL -POOL '75'
AND SUBSTRING('ADALOG-LINB-TEXT , 42 ,0 2 ) GT
..
ADD 1 'TEXT WRITJ: WRlTE WRITJ:
TO 'LC : . 'ADALOG-LINB-TEXT WOlUt FILI: 2 'ASTUUSIt WOlUt FrLl: 2 'OUTPUT-LINJ: WOlUt FILI: 2 'ASTUUSIt
'HWM :_ TRUl!:
END-Ir
.. -------POR '11 •
POR 'U
Print r.oapon . e eode err a y ----- - --------
0 TO 999 1 TO 255
9.5 Natural Monitor Programs
IF 'CN'l'('U,'12) GT 0 ~SS 'PNR/RSP/CN'l" INTO 'OUTPUT-LINB
*"
183
'11 'RSP('11,'12)
'CN'l'('11,'12)
'SEAT" B'
0900 .* ------------------------------------0910 . . -- Read SLOG r-.port. 2 end Li11 _t.ri ..
0920
.* -------------------------------------
.. --
0930 fI2 , . 1 0940 JlW2. JlEAO WORlt 2 'AOAJU!p-LXNE-2 0950 IF SUBSTlUNG('ADARBP-LINE-TJ:XT-2 , 1,6) _ 0960 'START-2 :_ TRI]II: 0970 0980 ESCAPE TOP /* (RW2.) 0990 ENll-IP 1000 Ir SUBSTlUNQ('AOAJU!P-LINE-TJ:XT-2,2,32) 1010 'ror individu.l rep1.y dbid/tok.n' 1020 BSCAPB BOT'l'a( (JlW2 . ) 1030 ENll-IF 1040 . . 1 0 5 0 l l ' ISTART-2 _ TRlJl! 1060 'ARAAY-2('X2) , _ 'AOAJU!P-LXNE-TEX'l'-2 1070 ADD 1 TO 1080 ENll-Ir 1090 1100 END-WORK /* (JlW2.)
'1:2
1110 ••
'DATA B'
188
9 Appendix
1120 r2 . nlR 'U • 1 TO 30 1130 Ir 'ARRAI-2 (U2) • ' 1140
ESCAPE BOTTOM (1'2 . )
1150 1160 1170
END-Ir :Ir *MACHINB-CLASS EQ 'MAUlFRAME' WlUTB (2) '1ülRAI-2(':I2)
1180
E.LSE
1190
WlUTB
1200
1210 1220 1230 1240 1250
(0)
'ARRAY-211I2)
END - Ir
END-raR /* (1'2 . ) •• ------------------------------------•• Compare both ar".y. •• ------------------------------------1'3 . l'OlI. 111 • 1 TO 30
1260 •• 1270 11' SUBSTRrNG('ARRAT-l(tIl) ,2,17) • 'Itama on 1'11. 1280 •• U· suaSTJUNG('ARRAI-l (lU) ,20,45) 1290
SUBSTRING('ARRAY-2('Il) ,20,45)
1300 1310
Ir *MACNINB-CLA$S EQ ' Ml\.INFRAHZ' WRlTE (2) NOTI'l'LZ 11 .• ' (13) I 'SLOQ i , .-pty'
1320 1330 1340
1350 1360 1370 1380 1390 1400 1410 1420
1430 1440 •• USO 1460 1470 1480
I
WlUTB (0) NOTITLZ /1 •• ' (13) I 'SLOG i , ampty' I ' . ' (13) END-Ir BSCAPB IIOTTOK (1'3.
BNI)-II' 11'
1490 1500 1510
1520 1530
BLSE WlUTE
1700
Ir
(0) NOT:ITLI: .mpti~'
SUBSTR%NG('ARRAY-1(111) ,20,45) GT ' SUBSTRING{'ARRAY-1{'I1) ,20 , 4 5 ) GT SUBSTR:ING{'ARRAY-2{'I1) ,20,45) Ir 'MACHINE-CLASS EQ 'MArNFRAMB' WlUTB (2) NOTITLB
11 ' .' (46) 1 ' SLOG i . bein
VIElf ST - MC- CALC - EDATZ (5'15 . 0) =-~ ENTITY (A5D)
D~
, ,-,- ««< ««< , , ,-,- ««< ««< , ,-,- ««< • ««< ,
LAST-CHGQ-DATE-TIMK (AIS) LAST-CHGD-USBR (AOS) (11.10) (AOS) (AOS) (ADe) (A250)
CALC-CQDE
!l1'P-DAn: Brl'-DATE-~
BXP-DATE ~,.
,, ,, , 'rum
1 REDZPINZ DDM
2 100M (&365)
'1023
roNC
(Al)
Re (114)
(Tl DATE (0) 2 TS (B8) ,~ (1'19)
, , .,
11040
0360
'''''
2 DB (N5)
0370 2 Re (14) 0380 1 14011 0390 2 I'UNC (14) 0400 2 eTX (Bl56)
DUO
2 TEXT (A) OYNAMl;C
0420
2 HASH
0430 0440 0450 0460 0470 0490 0490 0500 0510 0520 0530 0540 0550 0560 0570 0590 0590 0600 0610
(820)
1 'flASH (a20) 1 'COUNT (Pl0) END-DErnlll: * roRKAT SG-r * DErnlll: SuaRO!JT:lNE flASH-ilTN CALLNAT " USR10tON" '10tO ASS:lGN 14011. FUNC _ 1 CALLNAT " USR4011N" 'tOll ASS:lGN 14011. FUNC 2 ASS:lGN '1023.MS *CPU-TIMZ R. READ HULU-FBTCH ON DOM BI ISN ASSIGN "011. TEXT • 'DOM CALLNAT "USRtO llN " "011 END-READ ASS:lGN '4011.FUNC • 3 CALLNAT " USR4011N" 14011
,. ««
Count.u da not match ««< , (REI) END-Ir 11' '4011 HASH _ 'HASH TIIJ:N walT!: / T"'4011.HASH ' Ch.ck.um. match' (Gal) ELSE walT!: / T"'4011 . HASH '»»> Ch.cksuma da not .... tch ««< , (RZI) END-Ir
WRlTl! "DAT4U "TrME 'Endood' I "TIMD (ST . ) (ZH-99: 99: 99' . ' 9) 1050 END
1040
Figure 209:
Monitor - Natural Program Checksum
191
192
9 Appendix
9.5.8 0010 0020 0030 0040 0050 0060 0070 0080 0090 0100 0110 0120 0130 0140 0150 0160 0170
Compare Amount Fjelds ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• • VERIrr I'ILI: 99 Arl'ER ADARES • • 11/19/2004 ~ • 09/23/2008 ~, a.us.d multi-I!.tc:h 100 to the r_d • 09/24/2008 ~, ch. "'1. d ...."ltif:. tch •• inpu.t ....1". • 12/15 /2008 ~, a.us.d di.pl.y I!or .l.p.ed time .nd cpu time • cpu time tor TSO .nd b.tch only • 12/22/2008 ~, corrected. cOlq>utation f:or 'TIM!) (hh : _ : . . . . . ) • 01/05/2009 ~, c:h. n9ed "ith 'COUNTER • a.us.d f:i.lda LOG-COUNTER .nd 'LOG-COUNTER • 01/13/2009 ~ CHANGE REPORTING DATE TO IlAVE l'OUll DIGIT YEAR AND IN • FORHII.T TITT/tKfDD • • DEI'INZ DATA
'CTR
••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••
LOCI\Jo
OlBO 1 ~IB VIZW 01' ST-RI-CNTRB-LINZ 2 LOG-COUNTER OUO 0200 0210 0220 0230 0240 0250 0260 0270 0280 0290 0300 0310 0320 0330 0340 0350 0360 0370 0380 0390 0400 0410 0420 0430
OHO 0450 0460 0470 0480
OUO
2 POST -TAX- CNTRB-AMI' 2 PRB-TAX-CNTRB-AMT 2 DB- POST-TAX- CNTRB-AMT 2 DB-PRB-TAX-CNTRB-AMT 1 'POS T-TAX-CNTRB-AMT (P12 . 2) 'PRJ:-TAX-CNTRB-AMT (P12 .2) 1 'DB - POST-TAX-CNTRB -AMT (P12 . 2) 1 'DB-PRE-TAZ-CNTRB-AMT (P12 . 2) 1 'LOG-COUNTER (P1 5 ) (N4) 1 'MULTll'ETeH 1 'START-T:tKB (11.10) (11.10) 1 'START-DATE (N7) / . in unit. ol! 10 .... . 1 'CPU-TIME-HS 1 'CPU-TIME-HH (N2) (N2) 1 'CPU-TlME-HM 1 'CPU-TlME-SS (N3 . 2) (11.10) 1 'DATE10 END-DEI'INE • INPUT 'MULTlPETCB • ST . SET TIME • 'START-T IME : . 'TIME 'START-DATE :_ 'OAT4U • AT TOP 01' PAGE MOVZ 'OATU Ta '!lATEI0 I!XAKIHB PULL 'DATE10 '- ' RJ:i'LACB Wl:TH ' / ' WRlTE NOTITLI! NOIIDR 'PAGE: ' 'PAGE- NtlMBER 55T 'OATE10 'TlME (AL-8) END-TOpPAGI!: Rl . RD.D MULTI-FIlTeR IMULTIl"IITCR CONTRIB ADD POST-TAX-CNTRB-AMT TO 'POST-TAX-CNTRB-AMT ADD PRB-TAX-CNTRB-AMT Ta 'PRB- TAX-CNTRB-AMT ADD DB-POST-TAX-CNTRB-AMT TO 'DB-POST-TAZ-CNTRB-AMT ADD DB-PRB-TAX-CNTRB-AMI' Ta 'DB-PRB-TAX-CNTRB-AMI' ADD LOG-COUNTER TO 'LOG-COUNTER
,
0500 0510 0520 0530 0540 0550 0560 0570 0580 • 0590 WRITE ' CONTRIBUTION I'ILI: CHECIt SuteO.RI ' 0600 WlUTE 0610 2X ' POST-TAX-CNTRB-AHT ' 0620 25T 'POST-TAX-CNTRB-AMT (EM-ZZZ , ZZZ,ZZZ,ZZ9.99) / 2:.: ' PRIl:-TAX-CNTRB-AHT ' 0630
~--
193
9.5 Natural Monitor Programs
0640 0650 0660 0670 0680 0690 0700 0710 0720 0730 0740 0750 0760 0770 0780 0790 0800 0810 0820 0830 0840 0850 0860 0870 0880 0890 0900 0910 0920 0930 0940 0950 0960 0970 0980 0990 1000 1010 1020 1030 1040 1050 1060 1070 1080
25T 'PRE-TAZ-~-~ (EM-ZZZ,ZZZ,ZZZ,ZZ9 . 99) / 2X ' DB - POST -TAX -~ -ANT' 25T 'DB-POST-TAZ-~-~ (EM-ZZZ,ZZZ,ZZZ,ZZ9 . 99) / 2X ' DB - PRE -TAX- ~- ANT' 25T 'DB-PRE-TAZ-~-ANT (EM-ZZZ,ZZZ,ZZZ,ZZ9,99) / 2X ' LOG-COUNTER' 2
"
I I 3600
'CPU-TIME-~
'CPU-TlMB-~
('CPU-TlMB-HH
'CPU-TlMB-SS 'CPU- TIME - MS
'CPU- TlMB-SS 'CPU- TlMB - MS
('CPU- TlMB- SS
01T 01T 01T 01T 01T 01T
'Tima. ------------'Start Tima . . : 'END TDIZ . . , . : ' 'El.p•• d Time:'
('CPU-TIHE-~
'" '" '"
", ,---------- , '" '*foD/nyy' ---------- TlMB '" 'START- DATE '" fSTART '" *DAT, '"''l '"'" '"
,-
,,,,-
,. ,,-
195
'n/.' 'OPT'
'-'
'RPL' 'STATUS' 'TOB' 'OSSI'
,- ,----, ,- ,-------- , ,- '--- ' , ,- ,----
** ------------------------------------** -- Re.d lUJAREP report a nd J!'ill _t.rix ** ------------------------------------':Il :_ 2 RWl.. RBAD WOlUt 1 'ADAREP-LINB
** --
Ii' SUBSTRING I 'ADAREP-LlNE-TBXT, 1 , 15) _ '11. DAR E P 'rIRST-L:IHE : _ SUBSTRINQ('ADAREP-L:IHE-TEIT,1,80) ESCAPE TOP / * (RWl. ) END-Ir
V'
••11'
• •
•
8UBS~NQ('ADAREP-Lmz-'1'EX'1', 1,7) • '* pu. ' 'NBW-PILE ; . TRUE ADD I TO '11 HOVE SUBSTRING('ADAREP-LINB-TItXT, 10 , 3) TO 'i'ILE ('11 ) ESCAPE TOP /* (RWl.) END-Ir
Ir SUBS~NG('ADJUUI:P-LlNE-TItXT. 1 ,20) _ 'RPL ('11) , . 'HS' J:SCAPB TOP /* (RWl . ) END-IP
'ril. i .
r.plic:at..d :
Ir SUBSTRING('ADAREP - LINZ- TEXT,1,41) Replic:at.ion i . inact.i". J!'or thi. J!'il. ' 'ACT- lNACT ('11) , . 'inacti. . ' J:SCAPB TOP /* (RWl . ) END- lI' 11'
8UBS~NG('ADAREP-LINZ-TEXT,1,23)
Raplicat.or tarqat. ID ' 'TOB ('11) ; . SUBSTRINQ('ADAREP-LINZ-TI!XT,tl,4) ESCAPE TOP /* (RWl.)
•
•
END-lI'
Ir
SUBS~NQ('ADAREP-LINE-TEXT,1,44)
Ir
SUBS~NG('ADJUUI:P-LINE-TItXT.1,4")
Coll.ct bel!or. imeqe. ol! updete. 'OSSI ('11) :_ ' On' ESCAPE TOP /* (lun.) END-Ir
Coll. ct bal!or. imeqe. ol! updete. 'D8B:I (':Il) :_ 'OJ!'J!" ESCAPE TOP /* (RWl.) END-Ir END - WOR1t /* (RWl . )
_ r •• '
_ No
••
* * ------------------------------------------** -- Writ.. raport. out. oJ!' _trix
** ••
-------------------------------------------
1 1' *MACBINZ-CLASS EQ 'MAINJ'RAME' WlUTE (1) NO'l'ITLE NOHDR OlT ' - ' (78)
/
OlT *DATN (EM-9999-99-99) *TIME (AL-8) 30T ' Adaba. Replicator Report' 68T ' Paqe ' *PAQ!I:-NUMBJ:R (1)
/
'
196
9 Appendix
OlT *l'ROGRAM
68T *LIBRARI-ID
I
OlT ' . ' (78)
I
.U.
OlT ,rIRST - LINZ (AL-79)
'WI'2-LINE : .
,-------------------------------------------------------------------
tflUTZ WOU FILl! 2 twr2-LINl!
OlT *DATN (EN-9999·9g·gg) .TIME (AL-8) 30T 'Adab •• Replicator Report' S8T
.PAGE-NUMBER (0)
'P.~·
I
OIT *PROORAM
'" I
seT *LIBRARI-ID OIT '·'(78)
I
'WI'2-LlNE : _ 'FIRST-LlNE WRI'l'B wau FILE 2 fWP2-LlNE 'WF2-LUlE : _
,------------------------------------------------------------------fWP2-L:INE
tflUTE WORit !'noB 2
'WF2-LlNE : . '
,
END-Ir
••
PI. pop. In • 1 TO 51l1l1 U' 'rIU('U) • ' , ESCAPI! BO'l'TOM (PI.)
Ir *NACHINZ-CLASS EI;l 'IO.INFRANI!:' WRI'1'B
(1)
OIT 'FILZI'Il)
OST '1U'l.('Il)
lOT 'ACT-lNACT (tU) lOT 'TDB (tU) 2ST 'OSBI (,n)
BLSS U'ILI!(1l1)
....,-~
twF2-SPACl:l
IWF2-RPL
twF2-SPACB2
1wr2-AC'l'-INAC'l'
fWF2-SPACB3
IWr2-TDB
1WP2-SPACE4
,
twF2-DSBI
.... ....
BN!;I-IF
END-raR , .
'" '" '" ,. ,. ,. ,.
'R1'LI,n) 'ACT-J:NAC'r ('U)
'=81'11) 'DSBI('U) ItIF2-LlNE
(rl,
Wri t. Work Pil. 3 with ADADBS comman~ ADADBS RBPLlCATION PILB-006,ON,TARGET-302,DSBI-on
r2. FOR '11 • 1 TO 999 IP 'RPL('Il) _ 'lES' AND 'ACT-INACT('Il) • 'inaetiva' COMl'RESS 'ADADBS REPLICATION PIL&-' 'nLJ:('U)
, ,ON, TARGI:T- '
'mB('U)
.
, ,DSBI-oPP' INTO twF3-0UTl'UT LEAVING NO SPACE
WRITE
1fC)RJI;
END-Ir END-FOR ,. P2 .
PILJ: 3 'WI'3-0UTl'UT
~
Figure 211:
Monitor - Natural Program Parse ADAREP for RPL Flag
9.5 Natural Monitor Programs
197
9.5.10 De!ermine Highes! ISN Curren!ly Used
· ---------------------------------------------------------------------• rSNHICH • Thi. i • • eommon module whieh tind. the hiqha.t rSII currantly • u .ad in tile .
• • • ---------• E.rly an • 20ll/01/28 • 20ll/03/25 • • • • • • • • • • O .~
Lewi. Pritch.rd N•• Module Die tar Storr adaptad tor a e1ient !.arry F r ."in moditied tc look tor . 11 tile. in a .paeitiad databa.e raqu •• tad by INPUT 8tet. . .nt. ooe. a direct call LF to dater.in. ti1e . xi.ta , and i t &xi.t. tollow. with • 2nd direct call Ll with '8' in CB-OOP2 tc q . t tha t i r . t unu •• d lSN, .ubtr.et 1 tram th. r •• ult. , and th. ra . u1t. i . th. 1 •• t u.ad 18N. Nota: i . not n ae •••• ri1y the hiqha . t lSN " . cord load..:!.
· --------------------------------------------------------------
OEPlNE DATA LOCAL 1 'H:IGHEST-:ISN (P10)
/
...................................................................... .
/. /. /.
/
ooPl11:1;1 FRCM USRl04lP LlaRARY SYSBrY Thi . proqr ... . erve • • • e xampl e how to da.iqtl a u .e r - datinad proqr... tc c.l1 ' USRl04lN' .
...................................................................... .
'-"'AL 1 CONTROL-BLOClt (ASO) 1 REIlEFnnr: CONTROL-BLOC!\. 2 CB-RESERVED (B02) 2 REDEFINE CB-Rl:8ERVED 1 FIRST-BYTE (BOl) 2 CB-(:MD (A02) (A04) 2 CB-CID 2 CB-FJ:I.J: (802) 2 REDEPINE CB-PILE 3 CB-DBID (BOl) (BOl) 1 CB-1"tIlI. 2 CB-R8P (B02) 2 REDEPINE CB-RSP 3 CB- DBID - ALT(B02) 2 CB-ISN (B04) 2 CB - lSL (B0 4) (S04) 2 ClI-ISO 2 ClI - PBL (B02) 2 ClI-RBL (B02) 2 ClI - SBL (B02) 2 CS-VBL (802) 2 ClI-IBL (B02) 2 CS-001'1 (A01) 2 ClI-OOP2 (AOl) 2 CS-AIlDl (AOS) 2 ClI-AIlD2 (A04) 2 REDE F INE CS-AIl02 3 CS-J\DD2-2X (A02) 3 CS-SUB-OODE(B02) 2 ClI-AIlD3 (AOS) 2 CB-J\DD4 (A08) 2 ClI-AIlD5 (AOB) 2 CB-CT (A04) (A04) 2 ClI-UA
,.
/. /" /. /"
H ' 30 ' tor 2 byte t i le. C - . . d eoda CODDand ID 2 byta t i le numbar
/" D. te b •• e 10 /. Pil . numba" /" Ra.pan.a eoda /
2 byta DBI0
/" /. /" /. /. /. /. /" /. /" /. /" /.
D.te b •• e 10 (altern.ta) ISII ISN lowar lim.1t rSH quant.1ty Po.,..t butte r l e nqth Racord butte " 1 e nqth Se.reh bu tte r lenqth Va1ue butte r 1enqth ISII butEe " 1 e nqth Command option 1 CODDand option 2 Addition. 1 Addition . 2
/" /. /" /. /" /.
8ub Coda r e.ult. Addition . 3 Addition , 4 Addition . 5 Command U U.e" ."e.
198
9 Appendix
1 I'QRMI\.T-lIU~ (AlOOD) 1 REDEPlNE PORMAT - BUFnR
,,,,,-
2 IQRMAT-ADOR
(7.01)
I Rll:CORD-B~ (ASOOO) 1 REDEPlNE RECORD-SU!'FBR 2 _CORD-ADDR
(A01)
I SKARCH-BUPTER (JUOOO) 1 REDBPINB SEARCH-SUl'FBR 2
SEARCH-ADDR
(A01)
I VALVE-BUIPTER (JUODO) 1 REDEFINE VALUB-BUFFER 2 VALUlhlUlDR (ADl) I ISN-BurnR (JUOOO) 1 REDEFINE ISN-BUFFBR 2 ISN-ADIlR (ADl)
I RESPONSE
(lOt)
'FlLE- NR
(N05) (A02) (N05) (NOS) (NOt)
,,, Ix ,,
IDDID
'l'tRST-I'ILE 'NO- OP- PIlZS
END-PBI'INE INPU'l'
'&nter DBID:' 'DBID 'St&rting PNR :' 'FIRST-PILE 'No . ot Lil •• : ' 'NO-OF-FILES
11' 'DBID _ 0 RIIIMPUT 'DBID 1 .
0, pI • • •• a ntar • non-,..ro v.lu.'
END-l I'
11' 'FIRST-PILE _ 0 'FIRST-PILE :_ 1 WRlTE 'Pir a t fil a . a . END-lI'
not
pro~i~d,
••• ~ a tart with fil a I'
11' 'NO-OF-PILES _ 0 'NO-OP-FILES : _ 10 WRlTE 'No. of fil • • •a. not provided, • • au.. naxt 1 0 fil •• fram l.t ' 'file will
~ ~ . pl.yed'
END-lI' MOVI! ALL K'OO'
1'0 CONTROL - BLOCK H ' 30' 1'0 J'lRST-EjYTl!: 'OP' 1'0 CB- CMD MOVI! 5000 TO ClI-QL MOVZ 1000 TO ClI-ISL ClI-FBL ClI-SBL ClI-VBL ClI-IBL MOVZ 'ACe- .' TO _CORD-SurFER. RESET CB-ADDI CB-ADD2 ClI-ADD3 ClI-ADD4 ClI-ADD5 PERFORM CALL-ADABAS
*DA'l'41 ' - ' .TIME '- Roopo",t: ot S1.OO Fil. nrro 'INI'O-LINE WRITl! WORJt PILE J UNI'O- LINE
JO OBJ12 -
••
FOR2 . roR tl • 1 TO 500 11' 'REPLINlCS('I) • ' ESCAPE BOTTOM
••
••
u . a~
END-l I' lCZAMlNE 'R&PLlNES(tI) FOR ' Racord. l "..dad ' GlVING 'FOI.INl:I4 1 1' .romm4 !i'l' 0 'INI'O- LINE ,_ SUBSTRING('REPLINESI'I ) , 4 , J2) WRITI! WORJt I'lLl: J ' INI'O-LlNE END-lI' BVoMlNE 'REPLINES (tU FOR ' ASSOI JJ90 Ull ' GlVING ' FOI.INl:I5
'
. PROQRAN
202
9 Appendix
U' 'lI'OUND5 GT 0 CONPRESS
'ur .pace a Uoe.tad (blk):' SUBSTRING ('PJ!PLlNES (U) ,11.08) '111: .pac. un" •• d (blk):' SUBSTRING l,gPLINES (U) ,63,08)
INTO
'INFO-LrKE
WRITI! WORlI: PILZ 3 fINFO-LINB
••
IDIII-II'
EXAKINJ: 'REPL:lNII:$(fI)
raR 'ASSOl 3no RlI'
GIVING 'FOUND6 I r 'FOUND6 GT 0 COMPRBSS
'NI . pace a llocat.d (blk):' SUBSTRING('REPLlNES('I) , 17,08) 'NX . pace "nu•• cl (blk):'
..
SUBSTRING('REPLlNES('I) , 63,08) INTO 'INWO-LINZ WRITI!: 1fORII> FILZ 3 'INFO-101NB ENIl-II' EXANINl! 'UPLlNES(fI) I'OR 'OATAI 3390 DS1' GIV1NG '1'OUND7 11' 'rot1ND7 GT 0 COMPRESS
'OS . pac.
alloe.tael (blk):'
SUlJSTRZNG ('REPLINB8 ('1:) ,17.08) 'OS .pac. unu •• d (blk):'
SUBSTRJ:NG('REPLlNES('J:) ,63,08) INTO 'INPO-LlNE WRITB 110_ ru.J: 3 'nO'O-L:INE
.. '* BND-II'
END-roll
Figure 213:
(1'OR2.)
Monitor - Natural Program RPL30REP - SUX:; Space Check
203
9.5 Natural Monitor Programs
9.5.12 READF30P - Read SLOG Content 0010 •• - - ----------------------------------------------------------------0020 •• READF30P : READ SLOQ, Li1. 30 oL ~ptor OB 312 0030 •• Writun by 1>1. t . r Starr - 14 O . . , _ r 2010 0040 0050 0060 0010 0080 0090 0 1 00 0110 0 1 20 0130 0140 0150 0160 0110 0180 0190 0200 0210 0220 0230 0240 0250 0260 0210 0280 0290 0300 0310 0320 0330 0340 0350
• • ------------------------------------------------------------------DEPlNE DATA LOCAL , . DB312 Hx 1 ST-SLOQ VIZW OF ST-SLOQ-312 2 AA-PIEUl AB-FIELD AC-FIEUl AD-FIEUl g-FIEUl Ar-FIEUl AG-FIEUl AB-FIEUl C·AN-FIBLD lW-I'IEUl (1: 10) AO-FU:Ul C·JIR-I'IBLD AR-I'IBUl (1 : 1 0) DA- I'IEUl C·SS-I'IBLD SS - I'IEUl (1 : 10) /" A2S0
,,, ,, ,, ,, ,, ,, ,,
, ,,, .ro ,, ,, ,,
(1'10) (N4) (N4) ~ (11.16) ~-"" 'I'ILE-COUNT (N4) 'READ-COUNT (N10) (ASO) 'STRING (NOS) (12) INIT 'DELAI
.".,.
END-DEFlNE •• 0360 AT TOP 01' PAGE 0310 WRITJ:
0380
/
0390
/
,
1'11. N _
, --------------------
I$N SLOQ'
--------
0400 BND-TOPPAGE 0410 ••
0420 R1. READ 0430
0440 0450 0460 0410 0480
ST-SLOG BI ISN STARTING FROH 1 EJt»aNE SS- I'"IBLD (2) I'OR ' ST - ' GIVING POSITION IN 'PO 11' 'PO GT 0 THIW MOVl! SUBSTRING(BB - I'IEUl(2) ,'PO) TO SS-QUT WRITE SS-QU'l' " ISN END - lI' ADD 1 TO 'READ-CQUNT
•••
0600 END-READ
0610 11'" "COUNTER (Ri . ) _ 0 0620
WRITE (0)
'No r . corcb Lound in SLOG Li1. 30 in DB 312'
0630 BLSIl 0640
WRITE (0)
'Nulab.r oL tran •• ction. in SLOQ:' 'READ-COUNT
0650 BND-II'" 0660 END
Figure 214:
Monitor - Natural Program READF30P - Read SLOC Content
204
9 Appendix
9.5.13 RPL·P002 - Compare File Component. 0010 •• 0020 • • EIIEN'l' REPLlCATOR ~ rfo~n .,. Proj a clo RPL- P{)02 , a.ad tha Adabaa a.port from tha ..... infr.... 0030
0040 0050 0060
0070 0080 0090 0100
OllO 0120
0130 0140
0150 0160
0170 0180
OUO 0200
0210
•• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• •• u
R40ad th. Ad.aba a a.port from tha •• rve"
Input :
c:MWU'Ol - R&AD WORJ; 1 (HBF02 - REAl) Jf()Rl!; 2
REPLlCATOR r . port • Hainf".... : ST . OBI!. . TI)(: . RPLI'MPMB , DDPRl:NT . nB0109 . OUTLIST CMilKF03 -
WlU'l'E WOR:K 3 -
EXCl!:L
fo~t:
GIlt tha valu •• tor DBJ:D
or
Datat> • ••
Fila ~C&1560 LCL: _ Surt LCL 0&11 Surt LCL call 15 :53:0.8.630 bo.156D LCL: p711b 00BE240o. 15:53:08.630. bo.1 5 6D LCL: (CVT-I NI 8BCDIC -) ASCII conveuion required 15:53,08.6301>&1560. LCL: (CVT-III) BYTE OIIDER conveuion nquired 15 :53:08. 630 bo.1560 LCL: (CVT-I N) ca in t.n'1th-80/0x5o. Buf'f.r-0_6Ba8 0.00.0: lD4oo51"2 UCC6UO 0.0110.0.0114 00.030._ OQÖODÜ· . . . . . . 11 . N2 . . . . •. 0010., 00.0000.0.00.0.000.0.00 o.2CBo.333 000.0000.0. ......... i!.3. 00.20: 0000.4040. 40404040 '0.40'0.40. 40404040 .. 11118118811181
.u....
.•.
9.6
wep Log Windows
207
0030: 40404040 40404040 40404040 40404040 nUU8UUUU8 0040: 00000000 00000000 00000000 00000000 15 : 53:08.630 b007S60 LCL: (CVT-Ilf) CB out x..ngtb-80!0"SO Buff.r-O_68a8 0000: 30'04&32 6066CCU ~OOO~'OO 01:0~0300 0 8H2 · tiO .. ~. D. .. . . . . . • . 0010: 0000000000000000 CB023303 00000000 ... i.3. 0020: 00002020 20202020 20202020 20202020 0030: 20202020 20202020 20202020 20202020 0040: 00000000 00000000 00000000 00000000 15:53:08 . 630 b007S~0 LCL: pCI> OOBU8A8 15:53:08.630 b007560 LCL: CB (IH) l.oM>gth_ BO!0,,50 Butt.raO_na8 +.- .....•. 0000: 30'OU32 ~066CCU ~OOOUOO 01:0~0300 0@H2·fiD .. ~.P. 0010: 00000000 00000000 ClI023303 00000000 • •1.3. 0020: 00002020 20202020 20202020 20202020 0030: 20202020 20202020 20202020 20202020 00'0: 00000000 00000000 00000000 00000000 15:53:08.630 b007560 LCL: .... ada 1d don. t.o CE TRUJ: lS:53:08.~30 b007560 LCL: DBID 251 DBltll"O.t ooiESr8C 15'53:08.630 b007S60 LCL' DBIO 251 toun
".BQ.
208
9 Appendix
15 ' 53:08.6301:>0.1560 LCL ' HO" eh. OI< RB/IB 15 : 53:08.6301:>0.1560 LCL : Input RB :t..nqtb_ 81g/0,,333 Bul!l!.r-0_6b.=3 0000, 00000000 0000005r OgC9nre r 9r3,.,r6 ... . ... _~o+ ö
. .. .. ... RI289316
0320, oooooon 1"il1"il1"31"il 1"~1"il1"11"2 1"il1"01"31"0 ... o,,-"'li0656a ... 20030~0120030 0330, r9l"3ro UOtI 930 1553,08 . 6301:>0.1560 LCL , lToUl .. 1619 , cb_ott..t .. 180, ""'_ b_lnq .. 1619 15 :53:08.6301:>0.1 5 60 LCL : AllAßA8 u •• r-qu. ue handl. _ 15 :53:08.6301:>0.1 5 60 LCL : vio .. rit.uu: DBID 2 5 1 15 , 53,08 . 6301:>0.1560 LCL , ... iow .. it.uu , . tU ch..:! butt... h • • 0.1 5 60 LCL : ... iowrit.uu, procoo . .inll ACB ro.qu ... t on OBIO 251 conv .. rtinq t.o ACH 15 :53,08.6301:>0.1 5 60 LCL : NP USRIPC walTE X (2 5 1-00BZ6010- 5 -0140n:60-0(-I609486212(124612U) 15 ' 53 : 08.6301:>0.1560 LCL ' v .. it.u.r, ti ..... l ... _ _ • • 0 15 :53,08.6301:>0.1 5 60 LCL : ... actu.r, OBID 251 15 : 53,08.6301:>0.1560 LCL : vior.aduu : DBIO 251 iIOYCnt 2 15 ' 53 : 08.6301:>0.1560 LCL , vior.actu ... , . tUch..:! butt. .. _0.1 5 60 LCL : ... ior.aduu : procoo. . ;'nq ACSX ro.qu ... t on OBID 2 5 1 conv .. rtinq I:>o.cJc t.o ACB 15 : 53,08.6301:>0.1560 LCL : NP lISRIPC RU.D X (251-00BE6010-2-0140rc90-0 I Ull24?6360) 15 : 53:08.6301:>0.1 5 60 LCL : vior.adu ... : ""Pr bul!l!. r 1:>o.c:1< tor an ACB/ACSX ro.quo. ... on WIlD 251 15 :53,08.6301:>0.1 5 60 LCL : con .... rt t.o aob : DBIO 25 1 15 : 53,08.630 1:>0.1560 LCL : l . n u ••;.bd- 8 15 :53,08.6301:>0.1 5 60 LCL : pdat pointinq t.o ti ... t AB butt. r addro. . . 02urllCB 15 :53,08.6301:>0.1 5 60 LCL : ca (IH) Lenlltb_ 80/0,,5 0 autt ..... O_68a8 0000, 30tOn32 6066CCU 90009tOO 08060300 O@H2 · tio . . . P. + .- .... .• . 0010, 0000000000000000 CB023303 00000000 ....... . i.3 . 0020, 00002020 20202020 20202020 20202020 0030, 20202020 20202020 20202020 20202020 OOtO, 00000000 00000000 00000000 00000000 15 : 53,08.6301:>0.1560 LCL : aob" to.tor. ""PY' ... po. 0 15 , 53,08 . 630 1:>0.1560 LCL , . eh . t ..... copy' . cI> ... poo 0 15 : 53,08.6301:>0.1 5 60 LCL : ....,.ivinq au' pac:Jcag. roopr ..... tinq an aeh acbl< c:on_raion r~ ... 15 : 53,08.630 ba1560 LCL : ACBX c.d c:oa. (H2) 15 ' 53 : 08.6301:>0.1560 LCL ' paIxI (02i9"68) 15 : 53,08.6301:>0.1 5 60 LCL : n_ rb (1) - n_ fb (1) 15 : 53,08.6301:>0.1560 LCL : abdi d (r) 15 ' 53:08.6301:>0.1560 LCL ' ABO dUIiIP :t..nqtb_U/Ox30 Butt. ... Ox2Utg68 0000, 30004732 4600UOO 00000000 00000000 0.021".1 . . 0010, ca020000 00000000 CB020000 00000000 t .. . ... . t. 0020, 00000000 00000000 00000000 F868BBOO ....... . ... . .,h" . • ... • ... • ... B . 15 : 53,08.6301:>0.1 5 60 LCL : 1"8 (out) :t..nqtb_ U/Ox40 Butl!.... Ox2UUc:8 0000, t1412C38 2C502Ctl 422C3530 2CU2CU ...... ,8 , P ,AB, SO , "','" . .. .. 1. 0010, U2C313 5 2C412C41 U2C382C 412C4U5 C,1 5 ,"',AD,B,"',ll.lI: ... . .. ( . 0020, 2C312C41 2C4U62C 312CU2C 40U2C31 ,1,"',111 , 1 ,"' , HA,l (. . 0030, 2C412C40 '22C3130 2C412C41 492C3130 ,"' , MB,10 , "',"'I,10 15 : 53,08.6301:>0.1 5 60 LCL : AllAßA8(rcv) UQ-handl . (0) 15 :53,08.6301:>0.1 5 60 LCL : ca (OUT) :t..nqtb_ 80/0x50 Bu tt.... Oxbo.6h& 0000 , 0040n32 6066CCU 90940000 0 8 060300 .@ H2 · tio .~ .. D. • +. • .a. 0010, 0000000000000000 CB023303 00000000 ....... . 1.3 . 0020, 0000202020202020 20202020 41010000 .... 0030: 20202020 20202020 20202020 20202020 0040, 00000000 00000000 00000000 00000000 ....... . . 15 : 53,08.630 ba1560 LCL : final rc: 0 ca 15 ' 53 : 08.6301:>0.1560 LCL ' (~-OUT) blCD1C -> AIlcn oon_ r d on ro.quir..:! 15 :53,08.6301:>0.1 5 60 LCL : (~-OUT) B1:TIC 0iID8R con ........ ion ro.qui.-.d 15 : 53 , 08.6301:>0.1560 LCL : (~-OUT) ca i n Length-80!0"SO Butf.r-O_68a8 0000, 0040n32 10UCCU 90940000 0 8 060300 . 8H2 ' tio. ~ .. D. + . . . .a. 0010, 0000000000000000 CB023303 00000000 ....... . i.3 . . 0020, 0000202020202020 20202020 41010000 .... 0030, 20202020 20202020 20202020 20202020 0040, 00000000 00000000 00000000 00000000 1553,08 . 630 1:>0.1560 LCL , (~-OUT) ca Out :t..nqtb_80!0.. 50 Bufh ... Oxbo.6h8 0000, 0040D51"2 UCC6UO 94900000 00030601: . 8UDit · N • • • • • • D H2 . .. -. . 0010, 0000000000000000 02ca0333 00000000 ....... . . i.3. 0020 , 0000404040404040 40404040 01410000 .... @U . . U@ . ... . 0030, 4040404040404040 40404040 40404040 8UU8UUUUU 0040, 00000000 00000000 00000000 00000000 15 ' 53:08.6301:>0.1560 LCL ' _ LCL oa11 tin h h..:! _ LCL oa11 tin h _ _ 15 : 53,08.630 1:>0.1 5 60 ~ : (D) - u.cB """ten .. Lenllth_ n/oxu Bul!t .. r-OxUOt .. c:O 0000, 01011"80004180001 CB023303 00000000 .. '; .... . E.3 . • .. .• . . . . • h o . 0010: 00003303 00000000 OOOOB\lSC 00000000 .. 3 .... . ... \. 0020, 60ll:AOOOO 02001800 CODC4200 CODC4200 . e . . ... . MiB .iiiB. ··1 · ··1 · 0030, C68m2.U 1:50C8606 09C501D3 00000000 u · -l.f .öA>6. r . . yV.t RBPL. 0040: 00000000 15 : ~3,08.630 1:>0.1 5 60 ~ : (D) Call DBID 2~1 - Loc:al Coapl . .... 15 : 53,08.6301:>0.1560 ~ : (D) _ bturni"9 tMD(ZBCDIC) , O ,,' dSt2'b~n • • : 0 ; Re , 0 _ 15 : 53,08.6301:>0.1 5 60 ca1linll Al. local _t.and..:l_oodaba . DB10-2 5 1 r c:-O
°
• 0'
-
'"
.
°
9.6
wep Log Windows
209
1$ ' 53:08.630 ~7560 t.oc.l~ . t \>IrIp o~ .,..,ly ACB block 0000, 0040D5 1"2 UCCU60 OOgOOOOO 00030_ . II ÖODi:f · ....... ~. 0010, 0000000000000000 02CB0333 00000000 •......... 1.3 .... • 0020: 000040'0 '040'0'0 '040'040 01410000 ' .. 8U UUU @.A .. • 0030, 40404040 40404040 40 404040 40404040 ' aIllUIIIUII9UIII@ ' 0040, 00000000 00000000 00000000 00000000 •........... . ....• 1$, 5 3,08 . 630 ~7!l60 t.oc.l~n \>IrIp o~ r.ply &AB block 0000, 0101n100 OUBOOO l CB023303 00000000 ' .. ,; ..... E.3 ..... • 0010, 0000330300000000 0000885C 00000000 • .. 3 .....•.. \ .... • 0020, 60u.oOOO 02001800 COOCUOO CODCt200 . ... . . .ÄfiB.ÄfiB. 0030, C6BU2A8 ~50C8606 ogCSD7D3 00000000 'Ai ' ·• . t.Ui.6 .... • 0040, 00000000 1$, 5 3,08.630 boI1 5 60 t.oc.l~ . t otrr : Ro.turned Rbl _ IU Ibl _ O 15,53,OB.630 boI7560 t.oc.U.• qu• • t otrr, Reply RB i . 0000: 00000000 00000051" D9C 9r2" I"U3nr6 • ....... O&c.uo-;.ö' 0010, 1"01"31"240 40404040 '0404040 4040 4040 ' ''ooe lllliIlIllIlU '
.
0320, 0000001"2 1'01"01'31"0 1"91"01'11"2 1"01"01'31'0 ' ... ~,,"~ • ,~ 0330, 1'"91'31"0 15, 53 ,08.630 boI1 5 60 _ _ _ _ _'.IORK UNU' DB10002 5 1 RJ:I'LY CLASSI'C_ _ _ __ 15 , 53,08.630 boi7560 Prooeu Cl u d eRep pQNE-O_9tfB 1$, 5 3,08 . 630 boI75 60 5EIIDIQ Mll:5SAGZ MsgRef.""noe-l)o.bcdc5lc Leru;lth- I44------------
------------RDA
15 , 53,08.630 boi7560 Buff • ....o 0000, 005100001 91"011:0000 •.... Y. 15,53,08.630 boI7560 Buff ..... l 0000, 4400000000000004 3158CBOO C6Bt:94B' ·0 ...... . 1[t . d~ . · 0010, 4B5A3030 31202020 UUUU 31 5 32020 ' KZ001 ADAN"IS 0020, 00000000 00000000 00000000 00000040 • ............... @' 0030, 033300000000009000000000 00000000 . 3 . . ..... • 1$, 5 3,08.630 boI1560 Buff ..... 2 0000, 0040D51"2 UCCU60 00900000 000306DI: . @ÖODit · . ...... p· 0010: 0000000000000000 02CB0333 00000000 • ....... .. 1..3 .... • 0020, 0000404040 404040 '0404040 01410000 ' .. UIIUIIU .A .. • 0030, 40'0'0'0 '040'040 40404040 40404040 ' ffUUUUUUU ' 0040: 00000000 00000000 00000000 00000000 • ................ • 15, 5 3,08.630 boI1560 OIUhndKoo. .aqeByll.t> Ro.f.rence-0xbcdtbcdc5lc ryp.-Poou h tent kept for r . u . ao. 15, 5 3,08.630 boI1560 I:nter Coadlock _ Oxbd0254 " - " ' u_ 2 l"roa,Choo TiOrg. tIf_ NIlTWOl\ll ADAN"IS T. r gootId-O,,1. 000003 Mf.r."c_ Oxbcdl C30368Fl 68r1rsro 68r1l'568 1"16BC16B 68E4 6BCl 68C21C668 Cl6BJ"'76B r'76BD168 016BC409 C3ICUBFl E66Br868 F86BC16B CCU6000 0000000 2 40404040 40404040 00000000 0000sm9
1"1 1"{1 68Cl r16BC168 E46BC3C( 68C3CUB C8681"168 68r86BCl 68C168C3 6BCl6BC3 CUBC3l>8 C31C268Fl C86Br168 r'76BD168 0768C4C3 C4C?6Br4 6Br46811:4 6BC16BC3 C?6BC3IC? C3IC9681"8 9000ROO CB033300 40404040 40404040 00000000 C9F21'8r9
kA\Ik~ilki.Iti..
iJelikaWköUkÄAJr. ~ilk1.J.hI.ilk
~~kil.(lJti
IGOr IT 15,53,08.630 .b82'0 >>>>>>>>>>>>>>>> llDA 0'UD-251 flBQUJl:ST «O/I 15:53:08.63041>5abO Al WOlUtING 0810-251 RJ:Q LOCAL_ _ _ __ 15:5l:08.630 41>8240 R.ooadSyncRouti"" DrvBlock_0l5abO Loca1P.equ1O.t: !!b1_SU .,,1_715 Sb1_0 V1>1-ll 11>1_0 15:53:08.630 4b5abO Locd!loOqU•• t IR: Ouap ot n 0000: C1Cl6Bn 6BD76BCl C26B1"5J'Q 6BC16BC1 · J.J.Jt"kxkAhOllkAkA· 0010: Cl6Brll"5 öBC1öBC1 C46Br8ö8 Cl6BC1C5 'Al<Mi~iliü' ~1
0310: 1'2 1"11'231: rDI"Al61P 360041"U 401COOOO · Oilo>ji>. .. 6.00H ... • 0320: 000000r2 I"OJ"OI"3J'Q r9J'Qr11'2 J"OJ'Q1"31"0 ' ... 00000000u"ilO/IIIOII · Ol30: "l"lJ"O .... 6-' 15'53'08.646 4b5abO L>II< , (I"IR$'f) thr . .4 block 00BD5168 15:53:08.ö46 41>5abO L>IIIIDREG13 RJ:$TQRII: CALLBR VALUE$ ~m
..• ------------------------------------------------------------------.. ------------------------------------------------------------------" U"
DBI"ZNITZONS, WORIt NID SAVE AAI!:AS
D
SAVBJIREA DS
""
'" ''"'
""Da "Da ""' " ~"' • , ,,,
STANDARD REGISTER SAVE ARD.
,.,~
D DDNAMlh.OO'l'PUT ,~PM, UlEct.-80 ,DSORG_PS D
,~
DSORG_PS,~R,EXLS~'!',DDNAME_ '.'.'.,.
..
Da~'
.jl!