Caché High Availability Guide

	Contents
Chapter 1:	Introduction

Chapter 2:	Write Image Journaling and Recovery
	2.1 Write Image Journaling 2.1.1 Image Journal 2.1.2 Two-Phase Write Protocol 2.1.3 Write Image Journaling Modes 2.2 Recovery 2.2.1 Recovery Procedure 2.3 Error Conditions 2.3.1 If Recovery Cannot Complete (UNIX and OpenVMS) 2.3.2 Sample Recovery Errors 2.3.3 Write Daemon Panic Condition 2.3.4 Write Daemon Errors and System Crash 2.3.5 Freeze Writes on Error 2.3.6 Responding to a Freeze 2.4 Limitations of Write Image Journaling
Chapter 3:	Backup and Restore
	3.1 Caché Backup 3.1.1 Setting Up a Backup 3.1.2 Backups are Concurrent 3.1.3 Creating Additional Backup Tasks 3.1.4 Deleting a Backup Task 3.1.5 Running Backups 3.1.6 Adding New Directories 3.1.7 Performing Multivolume Backups 3.1.8 Backup Log Files 3.2 Estimating Size of Backups 3.2.1 Running DBSIZE Interactively 3.2.2 Using the DBSIZE Extrinsic Function 3.3 Backup Considerations 3.3.1 Error Handling for Backups 3.3.2 Backup and Transaction Processing 3.4 Restoring from a Backup 3.4.1 Using the Backup History to Recreate the Database 3.4.2 Suspending Database Access During a Restore 3.4.3 Restoring Database Properties 3.4.4 Performing a Restore 3.4.5 Error Handling for Restore 3.5 UNIX Backup and Restore 3.5.1 Using UNIX Backup Utilities 3.5.2 cbackup Utility 3.6 OpenVMS Backup and Restore 3.6.1 Efficiency 3.6.2 Concurrent Operation 3.6.3 History Log 3.6.4 Using the OpenVMS BACKUP Utility 3.6.5 Using CBACKUP.COM 3.6.6 Restore on OpenVMS 3.7 Backing Up Selected Globals and Routines 3.7.1 Backup and Restore Utilities for Selected Routines
Chapter 4:	Journaling
	4.1 Overview 4.1.1 Database Integrity Protection 4.1.2 Global Journaling 4.1.3 Differences Between Journaling and Write Image Journaling 4.2 Journaling Configuration 4.2.1 Enabling Journaling at Startup 4.2.2 Caché Switches Journal Files Automatically 4.2.3 Configuring Journaling 4.2.4 Journaling Selected Globals 4.3 Uses of the Journal 4.3.1 Restoring a Database from the Journal 4.3.2 Automatic Journaling of Transactions 4.3.3 Rolling Back Incomplete Transactions 4.4 Caché Journaling 4.4.1 When Journaling is Required 4.4.2 The Journal File 4.4.3 Globals in Transactions are Journaled Automatically 4.4.4 Journaling on a Network 4.5 Journaling Tasks 4.5.1 Start Journaling 4.5.2 Stop Journaling 4.5.3 Switch Journal Files 4.5.4 Viewing the Journal File 4.5.5 Restoring the Journal 4.5.6 Displaying the Journal 4.6 Journaling Utilities 4.6.1 Use %NOJRN to Manage Journaling at the Process Level 4.6.2 Extrinsic Functions for Manipulating the Journal File 4.7 Journaling Limitations 4.7.1 Some Globals Usually are Not Journaled
Chapter 5:	Shadow Journaling
	5.1 Shadow Journaling Overview 5.1.1 Fast Transmission Mode 5.1.2 Compatible Transmission Mode 5.2 Enabling Shadow Journaling 5.2.1 Configuring the Source Database Server 5.2.2 Configuring the Destination Shadow Server 5.2.3 Setting Shadowing Properties 5.3 Using Shadow Journaling 5.3.1 Starting Shadow Journaling 5.3.2 Stopping Shadow Journaling 5.3.3 Purging Shadow Journal Files 5.3.4 Viewing Error Messages 5.4 Disaster Recovery Using the Shadow Server
Chapter 6:	System Failover Strategies
	6.1 No Failover 6.2 Cold Failover 6.3 Warm Failover 6.4 Hot Failover
Chapter 7:	Cluster Management
	7.1 Overview of Clusters 7.1.1 Cluster Master 7.1.2 Cluster Master as Lock Server 7.2 System Design Issues for Clusters 7.2.1 Determining Database File Availability 7.3 Cluster Configuration 7.3.1 Namespaces and Replication 7.4 Managing Cluster Databases 7.4.1 Creating Caché Database Files 7.4.2 Mounting Databases 7.4.3 Deleting a Cluster-Mounted Database 7.5 Caché Startup 7.6 Write Image Journaling and Clusters 7.7 Journaling on Clusters 7.7.1 Enabling Journaling 7.7.2 Shadow Journaling 7.8 Backup 7.8.1 Cluster Considerations 7.9 Restore 7.9.1 Automatic Rollback 7.9.2 Background 7.9.3 Special Considerations 7.9.4 Running JRNRESTO 7.9.5 Restoring the Journal 7.9.6 System and Cluster Event Log 7.9.7 Running the Cluster Restore 7.9.8 Known Issues for Cluster Journal Restores 7.9.9 Cluster-Mounted Databases and the JOURNAL Utility 7.10 Cluster Failover 7.10.1 Requirements for Successful Cluster Failover 7.10.2 Cluster Failover Process 7.10.3 Cluster Failed Error 7.10.4 Caché Startup or Cluster Mount During Failover 7.10.5 Conditions When Cluster Failover Does Not Work 7.10.6 Unsuccessful Failover Leads to Cluster Crash 7.11 Recovery 7.11.1 Cluster Crash 7.11.2 Recovery Uses Write Image Files 7.11.3 Recovery Process 7.11.4 When Recovery Completes 7.11.5 Unsuccessful Recovery 7.12 Cluster Application Development Strategies 7.12.1 Block Level Contention 7.13 Caché ObjectScript Language Features 7.13.1 Remote Caché ObjectScript Locks 7.14 DCP and UDP Networking
Chapter 8:	Caché Clusters on Tru64 UNIX
	8.1 Tru64 UNIX Caché Cluster Overview 8.2 TruCluster File System Architecture 8.2.1 Caché and CDSLs 8.2.2 Remastering AdvFS Domains 8.3 Planning a Tru64 Caché Cluster Installation 8.4 Tuning a Tru64 Caché Cluster Member
Chapter 9:	Caché and Windows 2000 Clusters
	9.1 Load-Balanced Cluster 9.1.1 Setting Up a Load-Balanced Cluster 9.2 Failover Cluster 9.2.1 Setting Up a Failover Cluster 9.3 Common Procedures 9.3.1 Create a Cluster Group 9.3.2 Create an IP Address Resource 9.3.3 Create a Generic Service Resource 9.3.4 Create a Generic Application Resource 9.4 Resource Properties 9.4.1 IP Address Properties 9.4.2 Controller Properties 9.4.3 Cube Properties
Chapter 10:	ECP Failover
	10.1 ECP Recovery Guarantees 10.1.1 In-order Updates Guarantee 10.1.2 ECP Lock Guarantee 10.1.3 Clusters Lock Guarantee 10.1.4 Rollback Guarantee 10.1.5 Commit Guarantee 10.1.6 ECP Recovery Guarantee 10.1.7 $INCREMENT guarantee 10.2 Cluster Failover with ECP 10.2.1 Client Fails 10.2.2 Server Fails 10.2.3 Network Is Interrupted 10.2.4 Cluster as an ECP Database Server