Oracle9i RAC Book by Mike Ault

Phone
800-766-1884

Oracle 11g Books

Oracle training

STATSPACK Viewer

ISBN 0-9727513-0-0 ISBN 13 978-0972751308	Library of Congress Number: 2003107664
622 Pages Perfect bind 9x7	PD 603
Shelving: Database/Oracle	Oracle In-Focus Series # 1

Oracle9i RAC
Oracle Real Application Clusters Configuration and Internals

Mike Ault & Madhu Tumma

Retail Price $59.95 / £37.95

Key Features	About the Authors	Table of Contents
	Reader Comments	Errata

Order now and get immediate online access to the code depot!

Only $41.95 (30% off)

Combining the expertise of two world-renowned RAC experts, Oracle9i RAC is the first-of-its-find reference for RAC and TAF technology. Learn from the experts how to quickly optimizer your Oracle clustered server environment for optimal performance and flexibility.

Covering all areas of RAC continuous availability and transparent application failover, this book is indispensable for any Oracle DBA who is charged with configuring and implementing a RAC clusters database.

Mike Ault is one of the world's most famous Oracle authors with over 20 books in-print, and Madhu Tumma is a recognized RAC clustering consultant.

Together, Ault and Tumma dive deep inside RAC and show you the secrets for quickly implementing and tuning Oracle9i RAC database systems.

* Presents a complete guide to the installation and configuration of Oracle9i Real Application Clusters.

* Describes secrets for successful design of RAC clusters

* Supplies expert internals of shared disk technology, raw devices and RAID with RAC.

* Exposes the internal concurrency, resource coordination and locking mechanism within RAC

* Explains configuration of Transparent Application Failover (TAF)

* Details tips for monitoring and tuning RAC applications
Furnishes tricks for using Oracle9i RAC Cluster Guard

About the Authors:

Mike Ault

Mike Ault has five Oracle Masters Certificates. Ault was the first popular Oracle author with his book "Oracle7 Administration and Management" and he also wrote several of the "Exam Cram" books.

Madhu Tumma

Madhu Tumma has been working as Software Developer, IT Manager, Database Administrator, and Technical Consultant for about 18 years. He has worked on a wide variety of projects and environments ranging from mainframe, client-server, eBusiness to managed services. He has provided consultancy to variety of clients on database clusters, business continuity and high availability solutions. His experience ranges across multiple relational database systems. Madhu is frequent speaker at Oracle World and IOUG where he presented many technical papers. Madhu has Master Degree in science and attended Business Management graduate program. He lives in New Jersey with his wife Hema and two children Sandi and Sudeep.

Table of Contents:

Chapter 1 - Introduction to RAC
The Dynamic Business Environment

The Explosive Growth of Business Data

Online and Real-Time Access to Corporate Data

Data Warehousing, Packaged Applications, E-Commerce Operations

Dependence on Database Management Systems

High Performance Systems

Introduction to Cluster Technology

Benefits of Real Application Clusters (RAC)

What is a DB Cluster?

Architectures of DB Clusters

What is a Cluster Interconnect?

Problems with Other Architectures

The IBM Shared-Nothing Configuration

Microsoft Federated Servers Database

Seeing the High Availability Spectrum

Real Application Clusters

Processing Prior to Cache Fusion

Oracle9i Real Application Clusters

Attributes of an Oracle RAC Cluster

Building an Oracle RAC Cluster

RAC Performance and Scalability

Chapter 2 - High Performance and Highly Available Databases

High Performance
Growth of Powerful Processors

Why Parallel Processing?

Opportunities for Parallelism

Scalability

Parallel Databases

Types of Parallelism

High Performance Computing Cluster – HPCC

Clusterize Applications

How Do You Clusterize Applications?

Highly Available Databases

Need for Highly Available Data

Failure

Availability

Reliability and Serviceability

Fault-tolerant Systems

Database Availability

Clustered Systems

Databases Issues

Oracle RAC – High Performance and Highly Available Database

Chapter 3 - Database Clusters
Overview of the Cluster Technology

How Clusters Differ from Distributed Systems

Clusters are Different from Fault-tolerant Systems

Database Clusters

Why Clusters?

Types of Clusters

Failover Clusters

Scalable High Performance Clusters

Application Server Clusters

Other Types of Clusters

Components of a Cluster

Cluster Nodes

Emerging Server Cluster Architectures

Cluster Interconnect

Essentials for Parallel Database Clusters

Cluster Interconnect Products

Infiniband Interconnect

ClusterWare

Concurrent Database Access

Failover Database Clusters

Resources, Resource Type

Resource Groups

The Concept of a Virtual Server

Failover Process

Examples

Failover Cluster Architecture

Oracle Database Service in HA Cluster

Issues with FO Clusters – Hidden Risks

Parallel Database Clusters

Shared-Nothing Model

Shared-Disk Model

Microsoft SQL Server Federated Database

IBM Offerings

Requirements for Parallel Clusters

Oracle’s Instance Membership Recovery

Cache Coherency and Lock Management

Chapter 4 - Real Application Cluster Architecture
Overview of Oracle Real Application Clusters

RMAN for RAC

RAC Workload Management

RAC Scalability

High Availability

Architecture

Oracle9i Basic Processes

Oracle9i Basic Internal Structures

Oracle9i RAC Additional Processes

Oracle9i RAC Additional SGA Areas

RAC Server Components

GCS Resource Modes and Roles

Accessing RAC

Chapter 5 - RAC Server and Disk Technology

Overview
Oracle9i RAC Disk System

RAW Partitions and Cluster File System

Raw Partitions

Cluster File System

Storage Technology and Redundancy

Storage Disk System

Understanding I/O Path

Host Bus Adapter (HBA)

Storage Redundancy Components

Multiple RAID Controllers and Storage Processors

Power Management

High Availablity with SAN

RAID and RAID Administration

Multiple Access Paths to Host

Logical Volume Manager

Cluster File System

Oracle9i RAC and CFS

Context Dependent Symbolic Link – CDSL

Veritas CFS

PolyServe Matrix Server (MxS)

HP Tru64 CFS

Oracle Cluster File System (OCFS)

Other CFS Solutions

Oracle Managed Files

Oracle Disk Manager

Server Redundancy

Necessity of Server Redundancy

Redundancy Features

Chapter 6 - RAC Installation and Configuration
Architecture and Components

Servers and Operating System

Interconnect

cluster_interconnect parameter

Cluster Software

Shared Storage

General Steps

Phase – 1

Phase – 2

Phase – 3

Sun Solaris

Hardware / System Details

Creating the Cluster Using Sun Cluster 3.0

Install the Volume Manager

Install the Clusterware (OSD) for RAC

Configure Shared Storage

Veritas Cluster Volume Management

UNIX Pre-Installation Steps

Using Veritas DBE / Advanced Cluster

Communication Stack

Shared Storage Configuration

VCS Framework and Service Groups

Veritas I/O Fencing and SCSI – 3 Persistency

Install DBE / AC and Configure Service Groups – First Stage

Install DBE / AC and Configure Service Groups – Second Stage

Using HP Tru64 UNIX

Hardware / System Details

Set up Disks

Creating Installation and Database Creation

Using HP – UX Clusters

Hardware / System Details

Creating the Cluster

Shared Storage Configuration

Configure HP ServiceGuard Cluster

UNIX Pre-Installation Tasks

Using IBM AIX

Hardware / System Details

What is VSD|?

Configure using HACMP / ES

General Parallel File System – GPFS

Oracle Installation

Using Windows Cluster

Hardware / System Details

Creating the Cluster

Configure Shared Storage

Using Linux

Hardware / System Details

Interconnect & Cluster Software

Shared Storage Volumes

Oracle Cluster File System (OCFS)

PolyServe Matrix Server

Oracle Installation and Creation of Database

Oracle Software Installation

Database Creation using DBCA

Database Creation Manually

Chapter 7 - Internals of Real Application Clusters

Overview of Cache Fusion

Evolution of Cache Fusion

Nature of Cache Fusion

Benefits of Cache Fusion

Concurrency and Consistency

Cache Coherency

Global Cache Service

SGA Components and Locking

SGA – System Global Area

Program Global Area (PGA)

Buffer Cache Management

What is a Dirty Block?

Multi – Version Consistency Model

RAC Components

Global Cache Service

Global Enqueque Service

Row-Level Locks

Global Resource Directory

RAC Processes

Resource Coordination

Synchronization

GCS Resource Modes and Roles

Concept of Past Image

Lock Modes

Block Access Modes and Buffer States

Cache Fusion Scenarios

Block transfers using Cache Fusion – Examples

Block Access, Grants, and Interrupts

Cache Fusion and Recovery

Recovery Features

Recovery Methodology and Steps

Recovery Process – Re-mastering Resources

Chapter 8 - RAC Administration
Parameter Management

Overview

RAC and Initialization Parameters

UNDO Management

UNDO Management in RAC

UNDO Tablespace Features

System Rollback Segment

Concept of Thread

Thread Features

Redo Thread Maintenance

Segment Space Management

Automatic Segment Space Management

Manual Space Management

Oracle Managed Files

Oracle Disk Manager

Server Control Utility

srvctl add

srvctl config

srvctl getenv

srvctl setenv

srvctl start

srvctl status

srvctl stop

GSD daemon

GSD Services

Using SQL*Plus

Starting the Database in Cluster Mode

Using Oracle Enterprise Manager

Configuration of OEM with RAC

Chapter 9 - Backup and Recovery
Overview of RAC backup and Recovery

Export

Cold Backup using Scripts

Hot Backups using Scripts

RMAN (Recovery Manager)

Third Party Solutions

Backup of RAC Database

Using RMAN for Backups

Backup Procedures RMAN and RAC

Recovery in the RAC Environment

Media Recovery in RAC Instances

Using RMAN to Recover a RAC Environment

Recovery in an OCFS Environment

Recovery in a Raw File System Environment

Parallel Recovery

Standby Databases in RAC Configuration

Setting up a Standby Database for a RAC Cluster to a Single-Instance (One Node)

Configuration when the Standby Database is a Single Node System

Cross-Instance Archival

Archive Log Gap Resolution and FAL

Chapter 10 - Failover with TAF
Overview of Transparent Application Failover

Load Balancing

Example listener.ora

Example TAF Configuration

TAF with Connect-Time Failover and Client Load Balancing

Configuring TAF to Retry a Connection

Configuring TAF for Pre-Establishing Connections

Verifying TAF Configurations

Using Instance Role for Configuring the Primary and Secondary Instance

Connection to Instance Role Type

Establishing a Connection to a Specific Instance

Using BACKUP with TSF Pre-Establishing a Connection

Using OCI Driver

Chapter 11 - Performance Monitoring and Tuning
Analysis of Performance Issues

Monitoring RAC Cluster Interconnect Performance

Use of the cache_transfer Views

Monitoring the GES Processes

Monitoring and Tuning using OEM

Configuring the Oracle Intelligent Agent with RAC

Step-by-Step Setup of the Intelligent Agents on UNIX

Monitoring RAC using OEM

Chapter 12 - Parallel Execution
Overview

Types of Parallelism

Parallelism in Oracle Relational Database

Parallel Execution Mechanism

Granules for Parallelism

Parallel Execution Servers

Degree of Parallelism – DOP

Parallel Query: SELECT

Parallel DML – Update, Merge, Delete

Parallel DML – Insert…Select

Parallel DDL Statements

Rules for Paralleling “Create Table as Select” (CTAS)

SQL*LOADER – Parallel Load

Other Parallel Operations

Initialization Parameters at a Glance

Monitoring and Diagnosing the Parallel Execution

Chapter 13 - RAC Design Consideration
Designing Equipment for Real Application Clusters

What are the Effects of Component Failure?

Provide Redundancy at Each Level

Designing for High Performance

Compartmenting Transactions to Specific Nodes

Creating Efficient RAC Data Objects

Proper Sequence Usage

Tablespace Design in Real Application Clusters

Extent Management and Locally Managed Tablespaces

Minimizing Table Locks to Optimize Performance

Performance for Object Creation in Real Application Clusters

Chapter 14 - Application Deployment
Overview

Database Consolidation

E-business Applications

RAC Suitability

Advantages of the RAC Database

Scalability

Unpredictable Loads

High Availability

Application Deployment Issues

OLTP Applications using ‘exclusive’ or Specific Data

RAC Suitability for Packaged Applications

SAP and Real Application Clusters

Siebel and RAC

Oracle eBusiness Suite

Middleware Software

JDBC Connectivity

WebLogic and RAC Database

Chapter 15 - Migration to RAC Databases
Overview

Migration Methodology

Moving a Single Instance to RAC

Upgrading Oracle 8.1.7 to 9.2.0.2

Moving HA Cluster to a RAC Cluster

Packaged Applications

In Case of SAP Applications

In Case of Oracle eBusiness Suite

Chapter 16 - Using RAC Cluster Guard
Examining the Basics of Cluster Guard Technology

Oracle Real Application Clusters Guard Packs

The RAC Guard PFSCTL Control Utility

The Real Application Clusters Guard Monitors

Oracle Real Application Clusters Guard Configuration Templates

PFSSETUP Utility

Concepts of Oracle Real Application Clusters Guard

Primary and Secondary Instance Roles

Preferred Primary and Secondary Nodes

Definition of the Home and Foreign Nodes

Architecture of Oracle Real Application Clusters Guard

Oracle Real Application Clusters Guard Packs

Resources

Using the RAC Cluster Guard Command Line Utility

Recovering from an Unplanned Outage on One Node

Recovering from Unplanned Outages on Both Nodes

Configuring the Call-Home Feature

Enhancing Application Failover with Role Change Notification

ORATAB File Entries for Oracle RAC Guard

Using dbms_libcache to Warm the Secondary Cache

Overview of Warming the Library Cache

Set Up of DBMS_LIBCACHE

Use of the dbms_libcache Package

Managing the Oracle Real Application Clusters Guard Log Files

What if a Failover Occurs While Datafiles are in Backup Model

Index:

.
.rhosts file

A
Acquisition Interrupt
Active/Passive architecture
address_list
Advfs file system
agentctl
alert_$oracle_sid.log
alter session force parallel ddl
alter system set
ARCH
async_write()
ault_primary
ault_secondary

B
Bladed Server
Block Arrival Interrupt
block pinging
Blocking Interrupt

C
cache coherency
Cache structure
CALLOUT Listener
cdslinvchk
cfscluster
CFSfsckd
CFSMount
CFSQlogckd
chown
CKPT
clu_add_member
clu_create
Cluster quorum disk
cluster_database
cluster_interconnect
clustercheck
ClusterPulse daemon
CLVM
cmapplyconf
cmcfg.ora
cmruncl
cmrunnode
Command Line Interface
Concurrent Logical Volume Manager
Concurrent Virtual Shared Disk
connect_data
consistent gets
Control Files
CQJ0
create persistent initialization
create()
cursor_sharing
CVMCluster
CVMVolDg

D
DAM
DAS
Data Access Method
data-block pinging
datafile_directory
data-partitioning model
db block gets
DB block PING
db_block_size
db_create_file_dest
db_create_online_log_dest_n
db_domain
db_file_multiblock_read_count
db_link
db_nk_cache_size
dba_tablespaces
dbca
DBCA
dbca_raw_config
dbms_defer_sys.schedule_push
dbms_libcache
dbms_space.free_blocks
dbms_space.space_usage
dboraslib
DBWR
dc_free_extents
dc_object_ids
dc_used_extents
Dedicated Server
DEFAULT buffer pool
description_list
DIAG
Dictionary Cache
Dictionary Cache Locks
Direct Attached Storage
DLM daemon
dlm_conflicts
dlm_requests
dml_locks
Dnnnn

E
Enqueues
Exclusive mode

F
failover_mode
fal_client
fal_server
freelist groups
freelists
fsck

G
gc_files_to_locks
General Parallel File System
GES/GCS Area
GLM
global cache converts
global cache cr block receive time
global cache cr blocks received
global cache current blocks received
global cache gets
Global clustering
Global Lock Manager
global_dbname
GPFS
GSD
gsdaemon_node_name.log
gsdctl
gsdctl stat
gv$cache_transfer
gv$cr_block_server
gv$sysstat

H
HA clusters
HBA
HBA drivers
High performance Computing Clusters
Hot spare
hot-plug component
hot-swap component

I
icx_parameters
Idle Standby
init.ora
installDBAC
instance_groups
instance_number
instance_role
inter-query parallelism
intra-query parallelism
ioctl

J
JAVA Pool
Jnnn

K
kaio()
KEEP buffer pool

L
Large Pool
large_pool_size
LCK
LCKx
LGWR
Library cache
Library Cache locks
lio_listio()
List structure
LISTENER
listener.ora
LMD
LMDx
LMON
LMSn
LMSx
load_balance
load_ocfs
local_listener
Lock structure
log_archive_dest_n
log_archive_format
log_archive_start
logical hosts
Logical unit number
Logical Volume Manager
lvcreate

M
main.cf
Mean Time Between Failures
Mean Time to Failure
Mean Time to Recovery
memb
memory channel interconnect
mkcdsl
mklv
MTBF
MTTF
MTTR
Multiblock Read Consistency
Mutual Takeover
mxmpio

N
NAS
net_service_name
NetApp Filer Cluster
netca
Network Appliance Filers
Network Attached Storage
Network Load Balancing cluster
Non-Uniform Memory Access
Null mode
NUMA

O
o_direct
o_directio flag
ocfstool
ocmargs.ora
ODM
odm_abort()
odm_commit()
odm_create()
odm_io()
oinstall
open()
OPS
ora_list
ora_oemagent_dir
Oracle Database Configuration Assistant
Oracle Disk Manager
Oracle Parallel Server
OracleServicesid
oracm
oraping_config
ORCLUdlm
osdba
osoper

P
PanPulse daemon
Parallel Cache Management
Parallel Server Patch
parallel_adaptive_multi_user
parallel_automatic_tuning
parallel_max_servers
parallel_threads_per_cpu
Parallelism
PCM
pctfree
pctused
pgsd_subsys
PMON
Pnnn
pread()
pre-empt and abort’
Psd drivers
public_name.san
pvcreate
pwrite()

Q
qlogckd
QMN

R
read()
readv()
RECO
Recovery Manager
recovery_parallelism
RECYCLE buffer pool
Redo log Buffers
Redo Log Files
remote_listener
Resource Group Manager
RGM
RMAN
rollback_segment
rootdg
Rotating Standby
row cache locks

S
SAN
SANPulse daemon
sccheck
scinstall
SCL daemon
scsetup
Segregated clusters
Server Parameter File
service_names
services.ora
Shared mode
Shared Pool
shared-disk
shared-nothing
skgxn
skgxp
smit cm_add_grp
smit configchk.dialog
smit group
smit hacmp
smit mkuser
smit user
smit vg
SMON
SMP
SMP architecture
Snnn
SPFILE
spfile.ora
Split-brain
spreport.sql
sqlnet.ora
srvconfig
srvctl add
srvctl config
srvctl getenv
srvctl setenv
srvctl start
srvctl status
srvctl stop
statspack.snap
Storage Area Networks
Symmetric Multiprocessors

T
Table locks
Tablespaces
tcp_recvspace
tcp_sendspace
tnsnames.ora
trace_enabled
Transaction locks
typeversion

U
udlm
uname
UNDO Tablespaces
undo_management
undo_retention
undo_suppress_errors
undo_tablespace
user_dump_dest
user_indexes
user_tables
useradd
utl_file

V
v$bh
v$cache_transfer
v$class_cache_transfer
v$enqueues_stats
v$file_cache_transfer
v$instance
v$obj_stats
v$pq_sesstat
v$pq_slave
v$pq_sysstat
v$px_process
v$rowcache
v$session
v$session_wait
v$sesstat
v$sgastat
v$system_event
v$temp_cache_transfer
varyonvg
vg_ops
vgcreate
vgextend
Virtual Shared Disk
vnode cache
volassist
Volume Manager Storage Administrator
Volume Manager Support Operations
VSD
vxdg
vxdiskadm
vxfen
vxfenadm
vxfenconfig
vxfencoorddg
vxfentsthdw
vxfsckd
vxinstall

W
write()

Errata:

Page 111

This page incorrectly states that "the multiple block size feature is not yet supported by Oracle9i RAC".

Page 341

In the Create Database statement, the filepath is missing a /

Also,

"To add an extra redo log thread:

ALTER DATABASE ADD LOGFILE THREAD 2

GROUP 4 ('u01/oracle/redo/file4a',

'u02/oracle/redo/file4a') size 50M,

GROUP 5 ('u01/oracle/redo/file5a',

'u02/oracle/redo/file5a', size 50M,

GROUP 6 ('u01/oracle/redo/file6a',

'u02/oracle/redo/file6a') size 50M"

There are two errors here:

1.) the path to the redo logs should begin with /u01 and /u02 respectively and not

u01 and u02.

2.) the second member of each group should be suffixed with a "b" and not an "a",

e.g) /u02/oracle/redo/file4b, /u02/oracle/redo/file5b, and /u02/oracle/redo/file6b.

Page 419

SELECT
   INST_ID,
   EVENT,
   P1 FILE_NUMBER,
   P2 BLOCK_NUMBER,
   WAIT_TIME
FROM
   V$SESSION_WAIT
WHERE
   EVENT IN ('buffer busy global cr', 'global cache busy', 'buffer busy global cache');

Should be:

SELECT
   INST_ID,
   EVENT,
   P1 FILE_NUMBER,
   P2 BLOCK_NUMBER,
   WAIT_TIME
FROM
   GV$SESSION_WAIT
WHERE
   EVENT IN ('buffer busy global cr', 'global cache busy', 'buffer busy global cache');

Reviews:
John from Atlanta, GA says:
Mike is my favorite Oracle author and I have all his books for the past ten years. They are all exceptional, and this one is no exception. RAC is difficult, and the example make it clear and easy to follow.

A reader from Louisiana says:
Oracle RAC is made simple in this book. Good thing about this book is, it focuses on the RAC and RAC related subjects. Many concepts of the clustering are dicussed in detail. It is a good buy and worth the money.

Laurence from Massachusetts says:
I am in the 10g beta program and I was happy to see that the authors kept their focus on only the 10g features that are important.

Many other new feature book want to cover all new features even if they are not important.

Mike Ault has done an excellent job with this book and all of the 10g features are explained in the context of how they will help improve database functions.

This book covers Oracle9i RAC, Oracle RAC, Oracle9i Real Application Clusters and Oracle Real Application Clusters.

Copyright © 1996 -2017 by Burleson. All rights reserved.

Oracle® is the registered trademark of Oracle Corporation. SQL Server® is the registered trademark of Microsoft Corporation.
Many of the designations used by computer vendors to distinguish their products are claimed as Trademarks