Oracle Data Guard 11gR2 Administration Beginner's Guide电子书

Oracle Data Guard 11gR2 Administration Beginner's Guide

Table of Contents

Oracle Data Guard 11gR2 Administration Beginner's Guide

Credits

About the Authors

About the Reviewers

www.PacktPub.com

Support files, eBooks, discount offers and more

Why Subscribe?

Free Access for Packt account holders

Instant Updates on New Packt Books

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Time for action – heading

What just happened?

Pop quiz – heading

Have a go hero – heading

Reader feedback

Customer support

Downloading the example code

Errata

Piracy

Questions

1. Getting Started

What is Data Guard?

Standby database

Physical standby database

Logical standby database

Snapshot standby database

Oracle Data Guard evolution

Version 7.3 – stone age

Version 8i – first age

Version 9i – middle age

Version 10g – new age

Version 11g – modern age

Oracle Data Guard architecture

Data Guard services

Redo transport services

Protection modes

Apply services

Redo Apply (physical standby databases)

Monitoring Redo Apply

Time for action – monitoring Redo Apply

What just happened?

Pop quiz – real-time apply consideration

SQL Apply (logical standby databases)

Role transitions

Switchover

Failover

Fast-start failover

User interfaces for administering Data Guard

SQL*Plus

DGMGRL

Enterprise Manager

Time for action – using interfaces to monitor Data Guard

What just happened?

Data Guard background processes

Other replication solutions and Data Guard

Storage-based replication solutions

GoldenGate and Streams

Summary

2. Configuring the Oracle Data Guard Physical Standby Database

Preconfiguration for Data Guard

Data loss consideration

Network bandwidth consideration

Preparing the primary database

Archive log mode

Time for action – enabling the archive log mode

What just happened?

Force logging

Time for action – enabling force logging

What just happened?

Standby redo logs

Time for action – configuring standby redo logs on primary

What just happened?

Fast recovery area (FRA)

Time for action – enabling FRA

What just happened?

Understanding initialization parameters

DB_NAME

DB_UNIQUE_NAME

LOG_ARCHIVE_CONFIG

LOG_ARCHIVE_MAX_PROCESSES

LOG_ARCHIVE_DEST_n

LOCATION and SERVICE

VALID_FOR

SYNC and ASYNC

AFFIRM and NOAFFIRM

COMPRESSION

MAX_CONNECTIONS

MAX_FAILURE

REOPEN

NET_TIMEOUT

DELAY

LOG_ARCHIVE_DEST_STATE_n

What just happened?

Creating the physical standby database

Standby database related initialization parameters

FAL_SERVER

STANDBY_FILE_MANAGEMENT

DB_FILE_NAME_CONVERT

LOG_FILE_NAME_CONVERT

The physical standby database instance

Time for action – starting the physical standby instance and making it ready for the RMAN duplicate

What just happened?

Using RMAN duplicate to create physical standby databases

Time for action – running an RMAN duplicate

What just happened?

Post-installation steps

Verifying the standby database configuration

Time for action – verifying the standby database configuration

What just happened?

Managing Redo Apply

Time for action – starting, stopping, and monitoring MRP

What just happened?

Verifying synchronization between the primary and standby databases

Time for action – verifying synchronization between the primary and standby databases

What just happened?

Time for action – testing real-time apply

What just happened?

Have a go hero – checking the network latency effect on real-time apply

Summary

3. Configuring Oracle Data Guard Logical Standby Database

Logical standby database characteristics

Not everything must be duplicated

Use for reporting at all times

Independent standby database objects

Protecting writes on replicated standby tables

Limitation for specific data types and objects

High availability and disaster recovery considerations

Preparation for the configuration

Time for action – checking for the unsupported data types

What just happened?

Time for action – searching for and fixing any table row uniqueness problem

What just happened?

Creating a logical standby database

Time for action – making a physical standby database environment ready for conversion

What just happened?

Time for action – converting a physical standby database into a logical standby database

What just happened?

Verifying the logical standby database

Time for action – checking the redo transport service status

What just happened?

Time for action – checking the SQL Apply service status

What just happened?

Have a go hero – check the services in a broken configuration

Customization and management in a logical standby database

Selective replication in a logical standby database

Time for action – working with skip rules on a logical standby database

What just happened?

Database Guard settings for the logical standby database

Time for action – changing the Database Guard setting

What just happened?

Disabling database guard for a session

Have a go hero – testing the NONE Database Guard mode

Creating objects on the logical standby database

Creating and re-creating tables

Creating scheduler jobs

Creating materialized views

Time for action – creating objects on the logical standby database

What just happened?

Have a go hero – skip, disable guard, insert, instantiate, and disable skip

Automatic deletion of archived logs

Deletion of the foreign archived logs

Files inside the fast recovery area

Files outside the fast recovery area

Deletion of the local archived logs

Summary

4. Oracle Data Guard Broker

Introduction to Data Guard broker

Data Guard broker features and benefits

Centralized and simple management

Cloud Control integration

Oracle Data Guard and RAC

Role transition with Data Guard broker

Data Guard fast-start failover

Recommendation

Data Guard broker components

Oracle Data Guard broker server-side components

Data Guard Monitor process (DMON)

Configuration file

Oracle Data Guard broker client-side components

DGMGRL utility

Enterprise Manager Cloud Control client

Implementation of Oracle Data Guard broker

Time for action – initial setup of Data Guard broker

What just happened?

Time for action – connecting to Data Guard broker

What just happened?

Time for action – basic monitoring with Data Guard broker

What just happened?

Management with Data Guard broker

Enabling and disabling broker configuration

Time for action – disabling broker configuration

What just happened?

Enabling and disabling a standby database

Time for action – disabling and enabling database

What just happened?

Changing configuration and database properties using broker

Have a go hero – more examples on property changes

Time for action – changing the database name

What just happened?

Changing the state of the database

Have a go hero – more examples on state changes

Troubleshooting Data Guard broker

Data Guard tracing

Most Common Data Guard broker issues

ORA-16797: database is not using a server parameter file

ORA-10458:standby database requires recovery

ORA-16737:the redo transport service for standby database "string" has an error

ORA-16715:redo transport-related property string of standby database "string" is inconsistent

ORA-12514:TNS:listener does not currently know of service requested in connect descriptor

Current listener description

Oracle Data Guard fast-start failover

Time for action – configuring fast-start failover

What just happened?

Troubleshooting observer configuration

Script to stop and start observer

Summary

5. Data Guard Protection Modes

The Maximum Protection mode

The Maximum Performance mode

The Maximum Availability mode

Choosing the correct mode for your requirements

Changing Data Guard protection mode

Time for action – changing the protection mode with SQL*Plus

What just happened?

Time for action – changing the protection mode with Data Guard broker

What just happened?

Have a go hero - protection mode transitions with DGMGRL

Time for action – changing the protection mode with Enterprise Manager Cloud Control

What just happened?

Pop quiz – precautions for primary database availability issue in the Maximum Protection mode

Summary

6. Data Guard Role Transitions

Role transition considerations

Switchover

Performing switchover with a physical standby database using SQL*Plus

Time for action – preliminary tests before performing switchover

What just happened?

Time for action – switchover with a physical standby using SQL*Plus

Performing switchover with a physical standby database using broker

Time for action – switchover with a physical standby using broker

What just happened?

Performing switchover with a physical standby database using EM Cloud Control

Time for action – switchover with a physical standby using EM Cloud Control

Performing switchover with a logical standby database using SQL*Plus

Time for action – switchover with a logical standby database using SQL*Plus

What just happened?

Pop quiz

Performing switchover with a logical standby database using broker

Time for action – switchover with a logical standby using broker

What just happened?

Failover

Performing failover with a physical standby database

Time for action – failover with a physical standby database using SQL*Plus

Have a go hero

Performing failover with a logical standby database

Time for action – failover with a logical standby using broker

What just happened?

Summary

7. Active Data Guard, Snapshot Standby, and Advanced Techniques

Oracle Active Data Guard

Why Active Data Guard?

Oracle Data Guard license

Enabling Active Data Guard

Time for action – enabling Active Data Guard if Redo Apply is running using SQL *PLUS

Time for action – enabling Active Data Guard if the standby database is shut down

Time for action – enabling Active Data Guard using broker

What just happened?

Monitoring Active Data Guard

From primary

From standby

Active Data Guard with applications

Active Data Guard with PeopleSoft

Time for action – Active Data Guard with PeopleSoft

Active Data Guard with EBS

Active Data Guard with TopLink

Active Data Guard with Oracle BI

Active Data Guard with SAP

Active Data Guard features

EXPDP from standby database using NETWORK_LINK (ADG)

Time for action – exporting a database backup from Active Data Guard

What just happened?

Time for action – using the ASH report from the standby database

What just happened?

Have a go hero – running Statspack from a standby database

Using a snapshot standby database

Time for action – converting to a snapshot standby database

What just happened?

Time for action – converting to a physical standby database

What just happened?

Have a go hero – convert the physical standby to a snapshot and vice versa using broker

Cascade standby databases

Limitations with cascade standby database

Time for action – cascade standby database

What just happened?

Advanced compression in Data Guard

Time for action – enabling advanced compression

What just happened?

Preparation of standby on a cross-platform Data Guard

Time for action – creating a cross-platform Data Guard setup

What just happened?

Data Guard tuning and wait events

Network tuning

Redo transport and apply tuning

Data Guard wait events

Summary

8. Integrating Data Guard with the Complete Oracle Environment

The Oracle Enterprise Manager Cloud Control integration

Time for action – adding the Data Guard configuration into Cloud Control

What just happened?

Have a go hero

Cloud Control Data Guard administration home page

Modifying the Data Guard configuration

Time for action – enabling/disabling fast-start failover

What just happened?

Have a go hero

Monitoring Data Guard performance

Using Incident Manager to monitor Data Guard

Time for action – setting the threshold and creating an incident for estimated failover time metric

What just happened?

RMAN integration

Integration requirements and best practices

Physical standby requirement

RMAN Catalog requirement

Using a different DB_UNIQUE_NAME

General RMAN best practices

RMAN settings for the Data Guard environment

Registering primary database in the catalog

Configuring RMAN settings for primary database:

Configuring RMAN settings for standby database

Checking the RMAN configuration

Time for action – recovering a primary database using a standby database disk backup

What just happened?

Have a go hero

Using block change tracking with Data Guard

RAC integration

A RAC primary database with a single instance standby database

A RAC primary database with a RAC standby database

Summary

9. Data Guard Configuration Patching

What is patch and what are patch types?

Interim patch

CPU/SPU patches

PSU patches

Patch set

Patching on Data Guard

What just happened?

Best practices of patching

Upgrading OPatch

Performing prerequisite checks of patch

How to clean up patch history?

What just happened?

Patching on Data Guard configuration

How to apply an interim/bug patch on logical standby?

Time for action – applying a patch on logical standby

What just happened?

How to apply a PSU patch on physical standby database using broker?

Time for action – applying PSU on a physical standby database

Pop quiz

What just happened?

How to apply patch set on physical standby (11.2.0.1 to 11.2.0.3)?

Time for action – patch set upgrade of physical standby

What just happened?

Have a go hero – in-place patch set installation

Summary

10. Common Data Guard Issues

Recreating the standby control file

Time for action – recreating the standby control file

What just happened?

Dealing with redo transport authentication problems

Time for action – changing the SYS password in a Data Guard environment

Pop quiz – the redo transport authentication problem in only one instance of the primary database

What just happened?

Time for action – changing the redo transport user

What just happened?

Dealing with UNNAMED datafiles

Time for action – resolving UNNAMED datafile errors

What just happened?

Have a go hero

Closing a gap with an RMAN incremental backup

Time for action – closing a gap with an RMAN incremental backup

What just happened?

Pop quiz – using a tape for SCN incremental backup

Fixing NOLOGGING changes on the standby database

Time for action – fixing NOLOGGING changes on a standby database with incremental datafile backups

What just happened?

Time for action – fixing NOLOGGING changes in the standby database with incremental database backups

What just happened?

Turning on Data Guard tracing

Have a go hero

Gathering diagnostic data

Alert log and trace files

Time for action – monitoring the database alert log using ADRCI

What just happened?

Data Guard broker logs

Dynamic performance views

Summary

11. Data Guard Best Practices

Configuring a connection failover

Transparent Application Failover (TAF)

Configuring the client-side TAF

Configuring the server-side TAF

Fast Connection Failover (FCF)

Time for action – configuring FCF for JDBC connections

What just happened?

Fast Application Notification (FAN)

What just happened?

The archived log deletion policy on the standby database

Time for action – the recommended configuration for archived log maintenance on a standby database

What just happened?

Using flashback on a standby database

Time for action – using flashback on a standby database

What just happened?

Database rolling upgrade using the transient logical standby database

Time for action – performing a rolling upgrade using the transient logical standby database

What just happened?

Have a go hero – one last switchover

Corruption detection, prevention, and automatic repair with Oracle Data Guard

DB_BLOCK_CHECKSUM

DB_BLOCK_CHECKING

DB_LOST_WRITE_PROTECT

Automatic block media repair

Summary

Pop Quiz Answers

Chapter 1, Getting Started

Pop quiz – real-time apply consideration

Chapter 5, Data Guard Protection Modes

Pop quiz – precautions for primary database availability issue in maximum protection mode

Chapter 6, Data Guard Role Transitions

Pop quiz

Chapter 9, Data Guard Configuration Patching

Pop quiz

Chapter 10, Common Data Guard Issues

Pop quiz – redo transport authentication problem in only one instance of primary database

Pop quiz – using tape for SCN incremental backup

Index