Oracle DBA Workshop

Introduction
 Contents
Architectural Components
Managing an Oracle Instance
Managing Control File
Creating Data Dictionary Views and Standard Packages
Managing Redo Log Files
ManagingTablespaces and Data Files
Storage Structure and Relationships
Managing Rollback Segments
Managing Tables
Managing Indexes
Reorganizing Data
Architectural
Components
 Objectives
After completing this lesson, you should be able to do the following:

Describe Oracle server architecture and its main components

List the structures involved in connecting a user to an Oracle instance

List stages in processing:

Queries
DML statement
COMMITs
 Overview
Instance

User Shared Pool

processes Library SGA
Cache

Data Dict. Data Redo Log

Cache Buffer Buffer
Server Cache
processes
PGA
SMON DBW0 PMON CKPT LGWR Others

Data Control
Parameter files files Redo Archived
file log files log files

Password
file
Database
Oracle Database Files

Data Control Redo

Parameter files files log files
file

Archived
log files
Password
file
Database
 Other Key Physical
Structures

Parameter
file

Archived
log files
Password
file
Database
 Oracle Instance

An oracle instance:
Is a means to access an Oracle database
Always opens one and only one database

Instance

Shared Pool
SGA Memory
structures
Data Redo Log
Buffer Buffer
Cache

SMON DBW0 CKPT LGWR Others

Background
PMON
processes
 Processing of a SQL
Statement
Connect to an instance using:
The user process
The server process
The Oracle server components that are used depend on the
type of SQL statement:
Queries return rows
DML statements log changes
Commit ensures transaction recovery
Some Oracle server components do not participate in SQL
statement processing
 Processing a Query

Parse:
Search for identical statement
Check syntax, object names, and privileges
Lock objects used during parse
Create and store execution plan
Execute: Identify rows selected
Fetch: Return rows to user process
 Shared Pool
Shared Pool
Library
Cache

Data
dictionary
cache

The library cache contains the SQL statement text, parsed code
and execution plan
The data dictionary cache contains table, Column, and other
object definitions and privileges
The shared pool is sized by SHARED_POOL_SIZE
 Database Buffer Cache

Data buffer
Cache

Stores the most recently used blocks

Size of a buffer based on DB_BLOCK_SIZE
Number of buffers defined by DB_BLOCK_BUFFERS
 Program Global Area
(PGA)
Not shared
Writable only by the server process
Contains:
Sort area
Session info
Cursor state Server
Stack space process

PGA
Processing DML Statement
User Shared Pool
process SGA

Data Redo Log

Buffer Buffer
UPDATE emp... Cache

4 1 2
Server
3
process
Data Control
Redo
files files
1 log files

Database
 Redo Log Buffer

Redo log
buffer

Size defined by LOG_BUFFER

Records changes made through the instance
Used sequentially
Is a circular buffer
 Rollback Segment
Old image

New
Table
image
Rollback Segment

DML statement
Before making a change, the server process saves the old data value into
a rollback segment. This before image is used to:
Undo the changes if the transaction is rolled back
Provide read consistency by ensuring that other transactions do not
see uncommitted changes made b the DML statement
Recover the database to a consistent state in case of failures
 COMMIT Processing
Instance

Shared Pool
1 SGA

Data Redo Log

Buffer Buffer
Server Cache
process 4
LGWR
3
2
User
process Data Control
Redo
files files
log files

Database
 Log Writer (LGWR)
Instance

Shared Pool
SGA LGWR writes when:
There is a commit
The redo buffer log
Data Redo Log
Buffer Buffer
Cache is one_third full
There is more than
LGWR 1 MB of redo
Before DBW0 writes
Data Control
Redo
files files
log files

Database
Other Instance Processes

Other required processes:

- Database Writer (DBW0)
- Process Monitor (PMON)
- System Monitor (SMON)
- Checkpoint (CKPT)
The archive process (ARC0) is usually created in a
production database
Database Writer (DBW0)
Instance

Shared Pool
SGA
DBW0
writes when:
Data Redo Log
Buffer Buffer
Cache There are many
dirty buffers
There are few
DBW0
free buffers
Timeout occurs
Data Control Checkpoint occurs
files files Redo
log files

Database
 SMON: System Monitor

Automatically recovers the instance

- Rolls forward changes in the redo logs
- Opens the database for user access
- Rolls back uncommitted transactions
Coalesces free space
Deallocates temporary segments
PMON: Process Monitor

Cleans up after failed processes by:

Rolling back the transaction
Releasing locks
Releasing other resources
 Archiving

Database archive mode

- NOARCHIVELOG for DBs that do not require recovery
after a disk failure
- ARCHIVELOG mode for production
ARC0 process
- Automatically archives online redo logs
- Preserves the record of all changes made to the
database
 Summary

In this lesson, you should have learned how to :

Explain database files: data files, control files, online redo
logs
Explain SGA memory structures: DB buffer cache, shared
SQL pool, and redo log buffer
Explain primary background processes:
DBW0,LGWR, CKPT, PMON, SMON, and ARC0
Explain SQL processing steps : parse, execute, fetch
Managing an Oracle
Instance
 Objectives

After completing this lesson, you should be able to do the following:

Create the parameter file
Start up an instance and open the database
Close a database and shut down the instance
Get and set parameter values
Manage sessions
Monitor the ALERT file and the trace files
 Overview
Instance
User server
process process Shared Pool
PGA SGA

ALERT
file

Trace
files
Data Control
Parameter files files Redo
file log files

Password
file Database
 The Initialization
Parameter File Instance

Shared Pool
SGA Library
cache
Data Redo log
buffer buffer Data dict.
cache cahe

SMON DBWR PMON CKPT LGWR ARCH

initU15.ora

SQL> CONNECT / AS SYSDBA

SQL> STARTUP PFILE=%ORACLE_HOME%\database\initDB01.ora
 Uses of Parameters

Size the System Global Area (SGA).

Set database and instance defaults.
Set user or process limits.
Set limits on database resources.
Define various physical attributes of the database, such as
the database block size.
Specify control files, archived log files, the ALERT file, and
trace file locations.
Parameter File Example
# Initialization Parameter File: initDB01.ora
db_name = DB01
control_files = (/DISK1/control01.ctl,
/DISK2/control02.ctl)
db_block_size = 8192
db_block_buffers = 2048
shared_pool_size = 52428800
log_buffer = 64K
processes = 50
db_files = 1024
log_files = 10
max_dump_file_size = 10240
background_dump_dest = /home/disk3/user15/BDUMP
user_dump_dest = /home/disk3/user15/UDUMP
core_dump_dest = /home/disk3/user15/CDUMP
rollback_segments = (r01, r02, r03, r04, r05, r06, r07, r08)
...
 Parameters That should
be specified
Parameter Description
BACKGROUND_DUMP_DEST Location where background process trace files are written
(LGWR, DBWn, and so on). This is also the location for the
alert log.

COMPATIBLE Version of the server with which this instance should be

compatible. The default is 8.1.0

CONTROL_FILES Names of the control files.

DB_BLOCK_BUFFERS Number of blocks cached in the SGA.

DB_NAME Database identifier of eight characters or fewer. This is only

parameter is required when creating a new database.

SHARED_POOL_SIZE Size in bytes of the shared pool.

USER_DUMP_DEST Location where user debugging trace files are created on

behalf of a user process.
 Commonly Modified
Parameters

Parameter Description
IFILE Name of another parameter file to be embedded within
the current parameter file. Up to three levels of nesting is
possible

LOG_BUFFER Number of bytes allocated to the redo log buffer in the SGA

MAX_DUMP_FILE_SIZE Maximum size of the trace files, specified as number of

operating system blocks.

PROCESSES Maximum number of operating system processes that can

connect simultaneously to this instance.

SQL_TRACE Enable or disables the SQL trace facility for every user
session.

TIMED_STATISTICS Enables or disables timing in trace files and in monitor

screens
Startup and Shutdown in
Stages
STARTUP OPEN
All files opened as
described by the control
MOUNT file for this instance.
Control file
opened for this
NOMOUNT instance.
Instance
started.
SHUTDOWN
SHUTDOWN
 STARTUP Command

Start up the instance and open the database:

STARTUP PFILE=%ORACLE_HOME%\database\initDB01.ora
 ALTER DATABASE
Command
Change the state of the database from NOMOUNT to
MOUNT:

ALTER DATABASE database MOUNT;

Open the database as a read-only database:

ALTER DATABASE database OPEN READ ONLY;

 Opening a Database in
Read-only Mode

Any database can be opened as a read-only database

A read-only database can be used to :
- Execute queries
- Execute disk sorts using locally managed tablespaces
- Take data files offline and online, not tablespaces
- Perform recovery of offline data files and tablespaces
 Shutdown Options

Shutdown Mode A I T N
Allow new connections X X X X
Wait until current sessions end X X X 0
Wait until current transactions end X X 0 0
Force a checkpoint and close files X 0 0 0

Shutdown mode:
X NO
A Abort I Immediate
T Transactional N Normal 0 YES
 Different Types of
SHUTDOWN
Transfer Shutdown Shutdown Shutdown
of Funds Normal Immediate Transactional
Check account
1 balances
Insert new Database
2 funds down
Remove funds
3 from old account
Database
4 Commit down

Database
5 Log out down
 Dynamic Performance
Views

Are maintained by the Oracle server and continuously

updated
Contain data about disk and memory structures
Contain data that is useful for performance tuning
Have a public synonyms with the prefix V$
 Accessing Dynamic
Performance Views
OPEN
Data dictionary
MOUNT
Dynamic performance views
reading data from disk
NOMOUNT
Dynamic performance
views reading from memory
SHUTDOWN
 Example
V$PARAMETER
V$SGA
V$OPTION
SGA V$PROCESS
V$SESSION
V$VERSION
V$INSTANCE

V$THREAD
V$CONTROLFILE
V$DATABASE
V$DATAFILE
V$DATAFILE_HEADER
Control file V$LOGFILE
 Example
Dynamic Performance View
Description
(accessible in the
NOMOUNT stage)

V$PARAMETER Contains information about the initialization parameters

V$SGA Contains summary information on the SGA

V$OPTION Lists options that are installed with the Oracle server

V$PROCESS Contains information about the currently active processes

V$SESSION Lists current session information

V$VERSION Lists the version number and the components

V$INSTANCE Displays the state of the current instance

 Example (Continued)
Dynamic Performance View Description
(accessible in the
NOMOUNT stage)

V$THREAD Contains thread information; for example, about the redo

log groups

V$CONTROLFILE Lists the names of the control files (Even though available,
this view returns no rows in NOMOUNT state.)

V$DATABASE Contains database information

V$DATAFILE Contains data file information from the control file

V$DATAFILE_HEADER Displays data file header information from the control file

V$LOGFILE Contains information about the online redo log files

 Displaying Current
Parameter Values
Use the command

SHOW PARAMETER control

Query the V$PARAMETER dynamic performance

view:

SELECT name FROM v$parameter

WHERE name LIKE %control%;
 Dynamic Initialization
Parameters
Some initialization parameters can be modified while
an instance is running.

ALTER SESSION SET SQL_TRACE=true;

ALTER SYSTEM SET TIMED_STATISTICS=true;

ALTER SYSTEM SET SORT_AREA_SIZE=131072

DEFERRED;
 Enable and Disable
Restricted Session

Use the STARTUP command to restrict access to a database:

STARTUP RESTRICT

Use the ALTER SYSTEM command to place an instance in

restricted mode:

ALTER SYSTEM ENABLE RESTRICTED SESSION;

 Terminating Sessions

Identify which session to terminate with the V$SESSION

dynamic performance view :

SELECT sid, serial# FROM v$session WHERE

username=SCOTT;

Execute the ALTER SYSTEM command:

ALTER SYSTEM KILL SESSIN 7,15;

 The ALERT File and the
Trace Files
Trace files can be written by server and background
processes.
The Oracle server dumps information about errors in
trace files.
The ALERT file consists of a chronological log of
messages and errors.
Server process tracing can be enabled or disabled by :
- An ALTER SESSION command
- The parameter SQL_TRACE
 Controlling the ALERT File and
the Trace Files
Instance
User server
process process Shared Pool
SGA

Trace files ALERT file Trace files

USER_DUMP_DEST BACKGROUND_DUMP_DEST
 Location of ALERT File
and the Trace Files

Initialization Parameter Description

BACKGROUND_DUMP_DEST Defines the location of the background trace file and ALERT
file

USER_DUMP_DEST Defines where trace files will be created at the request of the
users

MAX_DUMP_FILE_SIZE Specified in O/S blocks; Limits the size of user trace files,
not the ALERT file or background trace files.
 Guidelines

Check the ALERT file periodically to :

Detect internal errors (ORA-600) and block
corruption errors
Monitor database operations
View the nondefault initialization parameter
 Summary

In this lesson, you should have learned how to :

Create the parameter file
Start up and shut down an instance
Access dynamic performance views
Manage sessions
Describe the use of the ALERT file and the trace files
 Creating Data
Dictionary Views and
Standard Packages
 Objectives

Construct the data dictionary views

Query the data dictionary
Prepare the PL/SQL environment using the
administrative scripts
Administer stored procedures and packages
List the types of database event triggers
 Overview
Other objects created with the database:
Data dictionary System
- Base tables tablespace
- Views
Dynamic performance tables
Built-in PL/SQL packages
Database event triggers fire Other
automatically during a specific event, database
such as a server error. files
Database
 Data Dictionary
Central to every Oracle database
Describes the database and its objectives
Contains read-only tables and views
Updated by SQL commands: System
- DDL tablespace
- Some DML
Owned by the user SYS
Stored in the SYSTEM tablespace
Accessed with SELECT statements

Other
database
files
Database
 Data Dictionary Contents
The data dictionary provides information about:
Logical and physical database structure
Definitions and space allocations of objects
Integrity constraints
Users
Roles
Privileges
Auditing
Other information
 Base Tables and Data
Dictionary Views

The data dictionary contains two parts:

Base tables:
- Normalized
- Created with database using sql.bsq script
Data dictionary views:
- Used to simplify the base table information
- Accessed through public synonyms
- Created with the catalog.sql script
 How the Data Dictionary Is Used

The data dictionary has three primary uses:

The Oracle server uses it to find information about:
- Users
- Schema objects
- Storage structures
The Oracle server modifies it when a DDL statement
is executed
Users and DBAs can use it as a read-only reference
for information about the database
Data Dictionary View
Categories

DBA_xxx
All of the objects in the database

ALL_xxx
Objects accessible by the current user

USER_xxx
Objects owned by the current user
Data Dictionary Examples
General overview
- DICTIONARY - DICT_COLUMNS
Schema objects
- DBA_TABLES - DBA_TAB_COLUMNS
- DBA_OBJECTS - DBA_CONSTRAINTS
Space Allocation
- DBA_SEGMENTS - DBA_EXTENTS
- DBA_FREE_SPACE
Database Structure
- DBA_DATA_FILES
- DBA_ROLLBACK_SEGS
- DBA_TABLESPACES
 Dynamic Performance
Tables

Virtual tables
Information accessed from:
- Memory
- Control file
Synonyms begin with V$
DBA uses to monitor and tune the database
Listed in V$FIXED_TABLE
Example: V$DATAFILE for data file information
 Stored Program Units

PL/SQL
- Oracles procedural language extension to SQL
- Stored in the data dictionary
Java
- Stored in the data dictionary
- To execute, publish its call specification
External procedures
- Written in C
- Stored in a shared library
- To execute, publish the call specification
 Stored PL/SQL Program
Units

A procedure, function, package, or trigger

Can accept and return parameters
Functions can be called from SQL
A set of SQL and PL/SQL commands
Stored in the data dictionary
Loaded into the shared pool
 Stored PL/SQL Packages

A Package
Groups logically related PL/SQL types, items, and
subprograms
Has two parts:
- A specification describes its components
- A body implements the logic
Example: DBMS_SESSION.SET_ROLE
- Package: DBMS_SESSION
- Procedure: SET_ROLE
 Executing a PL/SQL Program Unit
SQL> EXECUTE dbms_session.set_role(.)

Instance
Server
SGA
Shared pool

DBMS_SESSION
System Control
tablespace files
SET ROLE
Databases BEGIN
END;
Other Redo log
data files files
 Package Specification and Body
EMP_PKG
package PROCEDURE hire...
specification
g_empno NUMBER;

PROCEDURE hire ...

EMP_PKG
v_ename VARCHAR2(80);
package
...
body
BEGIN ...

FUNCTION edit_empno ...

Oracle-Supplied Packages

DBMS_SESSION: Generates SQL commands such

as ALTER SESSION or SET ROLE
DBMS_UTILITY: Provides various utility routines
DBMS_SPACE: Provides segment space availability
information
DBMS_ROWID: Provides ROWID information
DBMS_SHARED_POOL: Keeps and unkeeps
packages in the shared pool
DBMS_LOB: Provides routines for operations on
BLOB and CLOB data types
 Obtaining Information
About Stored Objects
Data dictionary view DBA_OBJECTS:
- OWNER
- OBJECT_NAME
- OBJECT_TYPE
- STATUS:
- VALID
- INVALID
DESCRIBE command:
SQL> DESCRIBE dbms_session
 Troubleshooting

The status of dependent objects may be INVALID:

- When DDL commands are executed on referenced objects
- After creating the objects using the IMPORT utility
- When a dependent object is modified
For program units with an INVALID status:
- The Oracle server automatically recompiles
- The user can recompile manually
Creating Data Dictionary
Views

Script Purpose

Catalog.sql Creates commonly used data dictionary

views and synonyms
Catproc.sql Runs scripts required for server-side
PL/SQL
Administrative Scripts:
Naming Conventions

Convention Description

Cat*.sql Catalog and data dictionary Information

Dbms*.sql Database package specifications

prvt*.plb Wrapped database package code

utl*.sql Vies and tables for database utilities

Built-in Package Example
Package CREATE OR REPLACE PACKAGE
specification dbms_session IS
from
dbmsutil.sql PROCEDURE set_role
(role_cmd VARCHAR2);

CREATE OR REPLACE PACKAGE BODY

Package dbms_session WRAPPED
body from 0
prvtutil.plb abcd
abcd
abcd
abcd ...
 Triggers
Written in PL/SQL, Java, or C
Fire
- Automatically
- Implicitly
Stored in the database
Managed like other stored program units
Usage examples:
- Generate derived column values
- Enforce complex edits
- Log events
- Audit
- Gather statistics
 Parts of a Trigger
Triggering event
- Instance startup or shutdown
- A specific error message or any error message
- User logon or logoff
- A DML statement on a specific table or view
- A DDL statement on any schema
Restriction
- Optional boolean clause in a WHEN statement
- Trigger only fires if clause is TRUE
Action: Program unit block
 Trigger Example

CREATE OR REPLACE TRIGGER log_logon

AFTER LOGON ON DATABASE
WHEN (USER = SYS OR USER LIKE OPS$%)
BEGIN
INSERT INTO sys.event_log
VALUES (Logon || USER || at
|| TO_CHAR(sysdate, YYYY-MM-DD HH24:MT:SS));
COMMIT;
END;
/
Managing Control File
 Objectives

Explain the uses of the control file

List the contents of the control file
Multiplex the control file
Obtain control file information
The Use of the Control
File
Small, binary file
Data Control Redo
files files log files Required:
- At mount
- To operate
Linked to a single
database
Should be multiplexed
Loss may require
Database recovery
 Control File Contents

Database name and identifier

Database creation date
Data file and redo log locations
Tablespace names
Log history
Backup information
Current log sequence number
Checkpoint information
Multiplexing the Control
File
Control_files=(..\DISK1\control01.ctl,..\DISK2\control02.ctl)

Disk 1 Disk 2

Control01.ctl Control01.ctl
 Guidelines for Control
Files
You should:
- Multiplex the control file
- Include the full pathname in CONTROL_FILES
- Back up the control file after the database structure
changes
Control files:
- Are sized by CREATE DATABASE keywords
- Have a reusable section that can expand due to Recovery
Manager updates
 Obtaining Information

V$CONTROLFILE
V$CONTROLFILE_RECORD_SECTION
Performance views from the control file:
- V$DATAFILE
- V$TEMPFILE
- V$TABLESPACE
- V$LOG
- Others
 Summary

In this lesson, you should have learned how to:

Explain that the control file is required to mount and
operate the database
Multiplex the control file to avoid a single point of
database failure
Back up the control file after making changes to the
physical database structure
Managing Redo Log
Files
 Objectives

Explain the use of online redo log files

Obtain log and archive information
Control log switches and checkpoints
Multiplex and maintain online redo log files
Plan online redo log files
Troubleshoot common redo log file problems
Analyze online and archived redo logs
Using Redo Log Files

Data Control Redo

files files log files

Database
Redo Log Groups and
Members

Group 1 Group 2 Group 3

Disk 1
Member Member Member

Member Member Member Disk 2

 Oracle Architecture
SGA
Data buffer Redo log
Cache buffer

DBWn CKPT LGWR ARCH

Control
files

Parameter Redo
file Data Log Archived
log files
Password
file
Files files
 Without Archiving

Backup Disk failure

Data control
files files

50 51 100 101

11 12
 With Archiving

Archived redo logs Disk failure

Backup
50

Data control
files files 99

50 51 100 101

11 12
 Obtaining Information
About Archiving

SQL Command:
ARCHIVE LOG LIST;
V$DATABASE:
- NAME
- LOG_MODE
V$INSTANCE:
ARCHIVER
 Obtaining Information
About Groups

V$THREAD:
GROUPS
CURRENT_GROUP#
SEQUENCE#
Obtaining Information
About Groups and
Members
V$LOG:
GROUP#
MEMBERS
STATUS
SEQUENCE#
BYTES
Obtaining Information About
Groups and Members

V$LOGFILE:
GROUP#
STATUS
MEMBER
 Log Switches and
Checkpoints

Force log switches with the command:

ALTER SYSTEM SWITCH LOGFILE;

Control checkpoints with the initialization parameters:

- LOG_CHECKPOINT_INTERVAL
- LOG_CHECKPOINT_TIMEOUT
- FAST_START_IO_TARGET
Adding Online Redo Log
Groups

ALTER DATABASE ADD LOGFILE GROUP 3

(DISK3\redo0301.log)
size 1M;

redo0101 redo0201 redo0301

Group 1 Group 2 Group 3

 Adding Online Redo Log
Members

ALTER DATABASE ADD LOGFILE MEMBER

\DISK3\redo0102.log TO GROUP 1,
\DISK4\redo0202.log TO GROUP 2;

redo0101 redo0102 redo0201 redo0202

Group 1 Group 2
 How to Relocate or Rename
Online Redo Log Files

1. Copy the online redo log files to the new location

2. Execute the ALTER DATABASE RENAME FILE
command.
Dropping Online Redo
Log Groups

ALTER DATABASE DROP LOGFILE GROUP 3;

redo0101 redo0201 redo0301

Group 1 Group 2 Group 3

Dropping Online Redo
Log Members

ALTER DATABASE DROP LOGFILE MEMBER

\DISK3\redo0201.log;

redo0101 redo0201

Group 1 Group 2
Cleaning Online Redo Log
Files

Example:

ALTER DATABASE CLEAR LOGFILE

\DISK3\redo0101.log;
 Online Redo Log
Configuration

Group 1 Group 2 Group 3

Member Member Member

Member Member

Disk 1 Disk 2 Disk 3

 Possible LGWR Errors

One member of a group of two or more is not available

All members of the next group are not available
All members of the current group are not available
 Analyzing the Redo Log
Files
Track Changes:
- To the database
- To a specific table
- To a specific user
Map data access patterns
Undo changes to the database
Use archived data to perform tuning and capacity
planning
How to Use LogMiner

Specify UTL_FILE_DIR
Create a dictionary file

EXECUTE DBMS_LOGMNR_D.BUILD(v815dict.ora,
C:\ora815\admin\v815\log);
Specifying Log Files to Be Analyzed
Set up the V$LOGMNR_CONTENTS view:
Initialize a new list and specify the first log file.
EXECUTE DBMS_LOGMNR.ADD_LOGFILE(
c:\ora815\oradata\v815\redo01a.log,
DBMS_LOGMNR.NEW);

Add additional log files to the list.

EXECUTE DBMS_LOGMNR.ADD_LOGFILE(
c:\ora815\oradata\v815\redo02a.log,
DBMS_LOGMNR.ADDFILE);
Starting to Analyze the
Redo Log Files

Initialize a LogMiner session:

EXECUTE
DBMS_LOGMNR.START_LOGMNR(
DICTFILENAME=>
c:\ora815\oradata\v815\log\v815dict.ora);
 Tracking Changes to a
Table
View V$LOGMNR_CONTENTS to track changes for the EMP table:

SELECT timestamp, username, sql_redo

FROM v$logmnr_contents
WHERE seg_name = EMP;

TIMESTAMP USER SQL_REDO

----------------- -------- ----------------
14-APR-99 SYS update SCOTT.EMP set sal=..
14-APR-99 SYS update SCOTT.EMP set sal=..
14-APR-99 SYS insert into SCOTT.EMP ()...
Finishing Analysis of the
Redo Log Files

Finish the LogMiner session:

EXECUTE
DBMS_LOGMNR.END_LOGMNR;
 Obtaining Information
About Logs Being
Analyzed

V$LOGMNR_DICTIONARY
V$LOGMNR_PARAMETERS
V$LOGMNR_CONTENTS
 Summary

In this lesson, you should have learned how to

Explain the use of online redo log files
Obtain log and archive information
Control log switches and checkpoints
Multiplex and maintain online redo log files
Plan online redo log files
Troubleshoot common redo log file problems
Analyze online and archived redo logs
ManagingTablespaces and
Data Files
 Objectives

Describe the logical structure of the database

Distinguish the different types of temporary segments
Create tablespaces
Change the size of tablespaces
Allocate space for temporary segments
Change the status of tablespaces
Change the storage settings of tablespaces
Relocate tablespaces
 Overview

Control
files

Redo log
Data files files

Database
 Database Storage Hierarchy
Database

Tablespace Data file

Logical
Physical
Segment

Extent

Oracle OS block
block
 System and Non-SYSTEM
Tablespaces
SYSTEM tablespace :
- Created with the database
- Contains the data dictionary
- Contains the SYSTEM rollback segment
Non-SYSTEM tablespaces:
- Separate segments
- Ease space administration
- Control amount of space allocated to a user
 Creating Tablespaces
Create TABLESPACE app_data
DATAFILE ..\DISK4\app_data_01.dbf
SIZE 100M,
..\disk5\app_data_02.dbf
SIZE 100M
MINIMUM EXTENT 500K
DEFAULT STORAGE (INITIAL 500K
NEXT 500K
MAXEXTENTS 500
PCTINCREASE 0 );
 Space Management in
Tablespaces
Dictionary -Managed tablespaces:
- Default technique
- Free extents recorded in data dictionary tables
Locally Managed tablespaces:
- Free extents recorded in bitmap
- Each bit corresponds to a block or group of blocks
- Bit value indicated free or used
 Locally Managed Tablespaces
Create TABLESPACE user_data
DATAFILE ..\DISK2\user_data_01.dbf
SIZE 5OOM
EXTENT MANGEMENT LOCAL
UNIFORM SIZE 10M;

Reduced recursive space management

Reduced contention on data dictionary tables
No rollback generated
No coalescing required
 Temporary Tablespace
Used for sort operations
Cannot contain any permanent objects
Locally Managed extents recommended
UNIFORM SIZE = SORT_ARE_SIZE * N

CREATE TEMPORARY TABLESPACE temp

TEMPFILE ..\Disk2\temp_01.dbf
SIZE 500M
EXTENT MANAGEMENT LOCAL
UNIFORM SIZE 10M;
Changing the Storage Settings
ALTER TABLESPACE app_data
MINIMUM EXTENT 2M;

ALTER TABLESPACE app_data

DEFAULT STORAGE(
INITIAL 2M
NEXT 2M
MAXEXTENTS 999);
 Offline Status

Offline tablespace not available for data access

Some tablespaces must be online :
- SYSTEM
- Tablespaces with active rollback segments
To take a tablespace offline :
ALTER TABLESPACE app_data OFFLINE;
To bring a tablespace online :
ALTER TABLESPACE app_data ONLINE;
 Read -only Tablespaces

ALTER TABLESPACE app_data READ ONLY;

Tablespace only available for read operations

objects can be dropped from tablespace
TO create a ready-only tablespace on a worm
drive
- ALTER TABLESPACE READ ONLY;
- Move the data file to the WORM drive
- ALTER TABLSEPACE..RENAME DATAFILE..;
 Making a Tablespace Read -Only

The tablespace must be online.

In release prior to Oracle8i,No active transactions are allowed
Oracle 8i allows current transactions to complete
The tablespace must not contain active rollback segments
The tablespace must not currently be involved in an online
backup
 Dropping Tablespaces

Tablespace removed from data dictionary

Optionally,Contents removed from data dictionary
OS files not deleted

DROP TABLESPACE app_data INCLUDING

CONTENTS;
 Resizing a Tablespace

Change the size of a data file :

- Automatically
- Manually
Add a data file
Tablespace APP_DATA

App_data_02.dbf
100M

100M
App_data_o1.dbf App_data_03.dbf
100M 200M
 Enabling Automatic Extension of
Data Files
ALTER TABLESPACE app_data
ADD DATAFILE ..\DISK6\app_data_04.DBF
SIZE 200M
AUTOEXTEND ON
NEXT 10M
MAXSIZE 500M;
Tablespace APP_DATA

App_data_01.dbf 100M App_data_03.dbf 200M

App_data_02.dbf 100M App_data_04.dbf 200M

 Changing the Size Of Data Files
Manually
ALTER DATABASE
DATAFILE ..\DISK5\app_data_02.dbf
RESIZE 2OOM;

Tablespace APP_DATA

App_data_02.dbf App_data_03.dbf
100M 200M

App_data_01.dbf App_data_04.dbf
100M 100M
100M
 Adding Data Files to a Tablespace
ALTER TABLESPACE app_data
ADD DATAFILE ..\Disk5\app_data_03.dbf
SIZE 200M;

Tablespace APP_DATA

App_data_01.dbf App_data_02.dbf
100M 100M

App_data_02.dbf 100M
 Moving Data Files:
ALTER TABLESPACE
The tablespace must be offline.
The Target data files must exist

ALTER TABLESPACE app_data

RENAME
DATAFILE ..\DISK4\app_data_01.dbf
TO ..\DISK5\app_data_01.dbf;
 Moving Data files
ALTER DATABASE

The database must be mounted.

The target data file must exits.

ALTER DATABASE
RENAME FILE ..\\DISK1\system_01.dbf
TO ..\DISK2\system_01.dbf;
 Obtaining Tablespace
Information
Tablespace Information:
- DBA_TABLESPACES
- V$TABLESPACE
Data file information :
- DBA_DATA_FILES
- V$DATAFILE
Tempfile Information :
- DBA_TEMP_FILES
- V$TEMPFILE
 Guidelines

Use Multiple Tablespaces

Specify default storage parameters for tablespaces
Assign tablespace quotas to uses
Include directory path in filenames
Use MINIMUM EXTENT option
Use locally managed extents
Can have 1023 data files per tablespace
 Summary

In this lesson,you should have learned how to :

Use tablespace to separate data
Resize tablespace by :
- Adding data files
- Extending data files
Use locally managed tablespaces
Use temporary tablespaces
Storage Structure and
Relationships
 Objectives
Describe the logical structure of the database
List the segment types and their uses
List the keywords that control block space usage
Obtain information about storage structure from
the data dictionary
List the criteria for separating segments
 Overview
Database

Tablespace Data file

Logical
Segment
Physical

Extent

Oracle block OS block

 Type of Segments

Table

Table partition

Cluster
Index
 Types of Segments

Index-Organized table Index partition

Rollback segment
Temporary Segment
 Types of Segments

Lob Segment
LOB index

Nested
table
Bootstrap
Segment
Storage Clause Precedence
Oracle Default

Tablespace

Segment
 Extent Allocation and Deallocation
Allocated when the segment is :
- Created
- Extended
- Altered
Deallocated when the segment is :
- Dropped
- Altered
- Truncated
- Automatically resized(rollback segments only)
 Used and Free Extents

Data file

File header Used extent Free extent

 Database block : Review
Minimum unit of I/O
Consists of one or more OS blocks
Set by DB_BLOCK_SIZE
Set at database creation
Database Block Contents

Header
Free Space

Data
Block Space Utilization Parameters

INITRANS
MAXTRANS

PCTFREE
PCTUSED
 BLOCK Space Usage
PCTFREE=20 PCTFREE=20
Inserts
80 %
Inserts

1 2

80 %
Inserts
Inserts 40 %
4
3
 Data Dictionary Views

Used Extents Free Extents

DBA_EXTENTS DBA_FREE SPACE

Segments Data files

DBA_SEGMENTS DBA_DATA_FILES

Tablespaces DBA_TABLESPACES
 Querying DBA_SEGMENTS
General Information Storage Settings

- OWNER - INITIAL_EXTENT
- SEGMENT_NAME - NEXT_EXTENT
- SEGMENT_TYPE - MIN_EXTENTS
- - MAX_EXTENTS
TABLESPACE_NAME - PCT_INCREASE
SIZE
- EXTENTS Other Information
- BLOCKS - Location
- BYTES - Tuning
 Querying DBA_EXTENTS
Location
Identification
-
- OWNER
TABLESPACE_NAME
- SEGMENT_NAME
- RELATIVE_FNO
- EXTENTS_ID
- FILE_ID
- BLOCK_ID
SIZE
- BLOCKS
- BYTES
 Querying DBA_FREE_SPACE
Location
- TABLESPACE_NAME
- RELATIVE_FNO
- BLOCK_ID
SIZE
- BYTES
- BLOCKS
Organizing Tablespaces Based on
Fragmentation Propensity
Tablespace Usage Fragmentation
SYSTEM Data dictionary Zero
TOOLS Applications Very low
DATAN Data Segments Low
Indexn Index Segments Low
RBSn Rollback Segments High

Tempn Temporary Segments Very high *

 Summary
In the lesson ,you should have learned how to :
Use Tablespaces to :
- Separate segments to ease administration
- control users space allocation
Categorize segments by the type of information
stored in the segments
determine extent sizes using the storage clause
control block space utilization
Managing Rollback
Segments
 Objectives

Create rollback segments using appropriate

storage settings
Maintain rollback segments
Plan the number and size of rollback segments
Obtain rollback segment information from the data
dictionary
Troubleshoot common rollback segments problems
 Rollback Segment
Old image

New
Table
Image
Rollback Segment

Update transaction
 Rollback Segments:
Purpose

Transaction rollback

Read
Rollback segment
consistency
Transaction
recovery
 Read consistency
Table SELECT *
FROM TABLE

New image
Image at start of statement
 Types of Rollback Segments

SYSTEM: Used for objects in the SYSTEM tablespace

Non-SYSTEM: Used for objects in other tablespaces
- Private: Acquired by a single instance
- Public: Acquired by any instance
Deferred: Used when tablespaces are taken offline
with the immediate option
 Transactions and Rollback
Segments

1 2

4 3
Transaction 1 Transaction 2

Active extent Inactive extent

Growth of Rollback Segments

1 2 1 2

5 3
4 3
4

Active extent New extent

Inactive extent
Shrinkage of Rollback Segments

6 1 1 22
2
5
6 3
4 3

OPTIMAL

Active extent

Inactive extent
 Creating Rollback Segments

CREATE ROLLBACK SEGMENT rbs01

TABLESPACE rbs
STORAGE (
INITIAL 100K
NEXT 100K
MINEXTENTS 20
MAXEXTENTS 100
OPTIMAL 2000K );
 Bringing Rollback Segments
Online
Use the following commands to make a rollback
segment available

ALTER ROLLBACK SEGMENT rbs01 ONLINE;

Specify the following initialization parameter to ensure

rollback segments are brought online at startup:

ROLLBACK_SEGMENTS=(rbs01, rbs02)
How Instances Acquire Rollback
Segments
Acquire Compute
Are there
named required No
enough
private number
RBS ?
RBS of RBS

Yes
Acquire
public
rollback
segments

Bring all acquired

rollback segments
ONLINE
 Changing Rollback Segment
Storage Settings

Use the ALTER ROLLBACK SEGMENT command

Can Change OPTIMAL or MAXEXTENTS

ALTER ROLLBACK SEGMENT rbs01

STORAGE ( MAXEXTENTS 200 );
 Deallocating Space from
Rollback Segments

Use the ALTER ROLLBACK SEGMENT command

if extents are active, may not shrink to the
requested size

ALTER ROLLBACK SEGMENT rbs01

SHRINK TO 4M;
 Taking a Rollback Segment
Offline

Take a rollback segment offline to make it unavailable.

If transactions are using the rollback segment, the status is

temporarily changed to PENDING OFFLINE.

ALTER ROLLBACK SEGMENT rbs01

OFFLINE;
 Dropping Rollback Segments

A rollback segment must be offline before it can be

dropped

DROP ROLLBACK SEGMENT rbs01;

 Rollback Segments in the
Database
DBA_ROLLBACK_SEGS
Identification
- SEGMENT_ID
- SEGMENT_NAME
Location: TABLESPACE_NAME
Type: OWNER (PUBLIC or SYS)
Status: STATUS(ONLINE or OFFLINE)
Rollback Segment Statistics

V$ROLLNAME V$ROLLSTAT
USN USN
NAME EXTENTS
RSSIZE
XACTS
OPTSIZE
HWMSIZE
AVEACTIVE
STATUS
CUREXT
CURBLK
Rollback Segment: Current
Activity
V$SESSION V$TRANSACTION
SADDR SES_ADDR
USERNAME XIDUSN
SID UBAFIL
SERIAL# UBABLK
UBASQN
UBAREC
STATUS
USED_UBLK
USED_UREC
 Planning Rollback Segments:
Number
OLTP
- Many small rollback segments
- Four transactions per rollback segment
- Up to ten transactions per rollback segment
Batch
- Few Large rollback segments
- One per transaction
 Planning Rollback
Segments:
Number of Extents
0.50
Probability
0.40
of extending
0.30

0.20

0.10

0.00
0 10 20 30 40
Number of extents
 Rollback Segment
Problems

Insufficient space for transactions

Read-consistency error
Blocking session
Error in taking tablespace offline
 Insufficient Space for
Transactions
No space in tablespace
- Extend data files
- Allow automatic extension of data files
- Add data files

MAXEXTENTS reached for segments

- Increase MAXEXTENTS
- Re-create segments with larger extent sizes
 Read-Consistency Error
Table SELECT * Reused block
FROM TABLE

New image
Image at statement commencement
 Blocking Session
Blocking
session

1 1 2
4
5 3
3 2 4

Existing Extent
New extent
 Error in Taking a Tablespace
Offline
Cannot take tablespace containing active RBS offline
1. Determine which rollback segments are in the
tablespace
2. Take all of these rollback segments offline
3. Find active transactions using these rollback segments
4. Find the session ID and serial number
5. Terminate the session if necessary
6. Take the tablespace offline
 Summary

In this lesson, you should have learned how to :

List rollback segments usage:
- Rollbacks
- Read consistency
- Recovery
Create adequate rollback segments
Troubleshoot rollback segment problems
Managing Tables
 Objectives
Create tables using appropriate storage settings
Control the space used by tables
Analyze tables to check integrity and migration
Retrieve information about tables from the data
dictionary
Convert between different formats of ROWID
 Storing User Data

Regular table
Partitioned table

Index-organized table Cluster

 Structure of a Row

Database Block Row header

Column Length
Column value
 Oracle Data Types
Data type

User-defined Built-in

Scalar Collection Relationship

CHAR(N), NCHAR(N)
VARCHAR2(N) VARRAY
NVARCHAR2(N) REF
NUMBER(P,S) TABLE
DATE
RAW(N)
BLOB,CLOB
NCLOB ,BFILE
LONG,LONG RAW
ROWID,UROWID
 Data Types For Storing Large
Object
Long,LONG RAW LOB

Single column per table Multiple Columns per table

Up to 2 gigabytes Up to 4 gigabytes
SELECT returns data SELECT returns locator
Data stored in-line Data stored in-line or out -of -line
Supports object types
No object type support
Random Access to chucks
Sequential access to chucks
 ROWID Data Type
Unique identifier for a row
Used to locate a row

ROWID format

00000 FFF BBBBBB RRR

Data object Relative file Block Number Row Number

Number Number
 Restricted ROWID
Can identify rows within a segment
Needs less space

BBBBBBBB RRRR FFFF

Block Number Row Number File Number
 Collections

Collections are objects that contain objects.

VARRAYs are ordered sets of elements containing
account and a limit
Nested tables are tables with a column or variable of
the TABLE data type

Nested Table
VARRAY
 Creating a Table
CREATE TABLE employee(
id NUMBER(7),
last_name VARCHAR2(25),
dept_id NUMBER(7))
PCTFREE 20 PCTUSED 50
STORAGE (INITIAL 200K NEXT 200K
PCTINCREASE 0 MAXEXTENTS 50)
TABLESPACE DATA;
 Temporary Tables
The rows are private to the session

CREATE GLOBAL TEMPORARY TABLE employee_temp as SELECT *

FROM employee

Tables retain data only for the duration of a transaction or session.

ON COMMIT PRESERVE ROWS;

DML locks are not acquired on the data

DMLs do not generate redo logs
 Creating a Tables: Guidelines
Use a few standard extent sizes for tables to
reduce tablespace fragmentation.
Use locally managed tablespaces to avoid
fragmentation.
Use the CACHE clause for frequently used
,small tables
Row Migration and Chaining
Before Update After update

Pointer
 Changing Storage and Block
utilization Parameter

ALTER TABLE summit.employee

PCTFREE 30
PCTUSED 50
STORAGE(NEXT 500K
MINEXTENTS 2
MAXEXTENTS 100);
 Manually Allocating Extents

ALTER TABLE summit.employee

ALLOCATE EXTENT(SIZE 500K
DATAFILE ..\DISK3\DATA01.DBF
 Nonpartitioned Table
Reorganization
ALTER TABLE employee
MOVE TABLESPACE DATA1;

Moves data into a new segment while preserving

indexes,constraints,privileges ,and so on,on the
table
Is being used to move a table to a different
tablespace or to reorganize extents
 High-Water Mark
After Inserts:
Extents ID 0 1 2 3 4

After deletes : High-water mark

Extents ID 0 1 2 3 4

Used Block Unused Block Free space after

delete
 Finding The High-Water Mark
DBMS_SPACE.UNUSED_SPACE
TOTAL_BLOCKS

UNUSED_BLOCKS
Extents ID 0 1 2 3 4

High-water mark
LAST_USED_EXTENT_FILE_ID,
LAST_USED_EXTENT_BLOCK_ID
 Deallocation of Unused Space
Before deallocation
Extents ID 0 1 2 3 4

ALTER TABLE summit.employee DEALLOCATE UNUSED

After deallocation: : High-water mark
Extents ID 0 1 2 3 4

Used Block Unused Block Free space after

delete
 Truncating a Table
TRUNCATE TABLE summit.employee;
Extent ID 0 1

Free Space
High- Water
Mark
 Dropping a Table

Drop TABLE Summit.department

CASCADE CONSTRAINTS;
 Dropping a Column
Remove a column from a table

ALTER TABLE employee

DROP COLUMN comments
CASCADE CONSTRAINTS CHECKPOINT 1000;
This removes the column length and data from each
row,freeing space in the data block.
Dropping a column in a large table takes a
considerable amount of time
 Using the UNUSED option
Mark a column as unused

ALTER TABLE orders

SET UNUSED COLUMN comments
CASCADE CONSTRAINTS;

Drop unused columns

ALTER TABLE orders

DROP UNUSED COLUMNS CHECKPOINT 1000;

Continue to drop Column operation

ALTER TABLE orders

DROP COLUMNS CONTINUE CHECKPOINT 1000;
 Retrieving Extent Information
DBA_EXTENTS

OWNER
SEGMENT_NAME
EXTENT_ID
FILE_ID
BLOCK_ID
BLOCKS
 Summary
In this lesson,you should have learned how to :

Create a table with appropriate storage and

block
utilization parameters
Control table storage
Use the DBMS_ROWID package
Managing Indexes
 Objectives
List the different types of indexes and their uses
Create B-tree and bitmap indexes
Reorganize indexes
Drop indexes
Get index information from the data dictionary
Classification Of Indexes

Logical
- Single column or concatenated
- Unique or nonunique
- Function-based
Physical
- Partitioned or nonpartitioned
- B-tree
- Normal or reverse key
- Bitmap
 B-Tree Index
Index entry
Root

Branch

Leaf
Index entry header
Key column Length
Key Column Value
ROWID
 Reverse key index

Key ROWID
ID FIRST_NAME JOB
ID (BLOCK ROW# FILE #)
7499 ALLEN SALESMAN
1257 0000000F.0002.0001
7369 SMITH CLERK
2877 0000000F.0006.0001 7521 WARD SALESMAN
7566 BLAKE MANAGER
4567 0000000F.0004.0001
7654 MARTIN SALESMAN
6657 0000000F.0003.0001 7698 BLAKE MANAGER
7782 CLARK MANAGER
8967 0000000F.0005.0001
9637 0000000F.0001.0001
. .
9947 0000000F.0000.0001
 Creating Function-Based Indexes

CREATE INDEX summit.item_quantity_to_deliver_idx

ON summit.item(quantity-quantity_shipped);

Queries using expressions can use the index

SELECT ord_id, item_id

From ITEM
WHERE (QUANTITY -QUANTITY_SHIPPED) > 0;
Bitmap Index
Table File 3
Block 10

Block 11

Block 12

start end
Key ROWID ROWID bitmap
< Blue, 10.03, 12.8.3, 1000100100010010100>
<Green, 10.0.3, 12.8.3, 0001010000100100000>
<Red, 10.0.3,12.8.3, 0100000011000001001>
<Yellow,10.03,12.8.3, 0010001000001000010>
Comparing B-tree and Bitmap Indexes

B-Tree Bitmap

Suitable for high Suitable for low

- cardinality columns - cardinality columns

Updates on keys relatively Updates on keys columns very

inexpensive expensive

Inefficient for queries using OR Efficient for queries using OR

predicates predicates

Useful for data warehousing

Using for OLTP
Creating Normal B-tree Indexes

CREATE INDEX
summit.employee_last_name_idx
ON summit.employee(last_name)
PCTREE 3O
STORAGE(INITIAL 200K NEXT 200K
PCTINCREASE 0 MAXEXTENTS 50)
TABLESPACE indx;
 Creating Indexes :Guidelines
Balance query and DML needs
Place in separate tablespace
Use uniform extent sizes:Multiples of five block
or MINIMUM EXTENT size for tablespace
Consider NOLOGGING for large indexes
Set high PCTFREE if new key values are likely
to be within the current range
 Create Reverse Key Indexes
Create UNIQUE INDEX summit.orders_id_idx
ON summit.orders(id) REVERSE
PCTFREE 30
STORAGE(INITIAL 200K NEXT 200K
PCTINCREASE 0 MAXEXTENTS 50)
TABLESPACE indx
 Creating Bitmap Indexes
Use the parameter CREATE_BITMAP_AREA_SIZE to
specify the amount of memory allocated for bitmap
creation.

CREATE BITMAP INDEX ORDERS_REGION_id_idx

ON summit.orders(region_id)
PCTFREE 30
STORAGE(INITIAL 200K NEXT 200K
PCTINCREASE 0 MAXEXTENTS 50)
TABLESPACE indx;
Changing Storage Parameters for
indexes

ALTER INDEX
summit.employee_last_name_idx
STORAGE (NEXT 400k MAXEXTENTS 100);
Allocating and Deallocating Index
Space

ALTER INDEX summit.orders_region_id_idx

ALLOCATE EXTENT(SIZE 200K
DATAFILE ..\DISK6\INDX01.dbf);

ALTER INDEX summit.orders_id_idx

DEALLOCATE UNUSED;
 Rebuilding Indexes
Use the ALTER INDEX Command to:

Move an Index to a different tablespace

Improve space utilization by removing deleted entries
Change a reverse key index to a normal B-Tree
index and vice versa

Alter index Summit.orders_region_id_idx REBUILD

TABLESPACE indx02;
 Online Rebuild of Indexes
Rebuilding indexes can be done with minimal table
locking

ALTER INDEX summit.orders_id_idx REBUILD

ONLINE
 Coalescing Indexes

Before Coalescing After Coalescing

ALTER INDEX summit.orders_id_idx COALESCE

 Dropping Indexes
Drop and Re-create an index before bulk loads
Drop indexes that are infrequently needed and
build
them when necessary
Drop and re-create invalid indexes.

DROP INDEX
summit.department_name_idx;
Obtaining Index Information

DBA_INDEXES DBA_IND_COLUMNS
OWNER INDEX_OWNER
INDEX_NAME INDEX_NAME
INDEX_TYPE TABLE_OWNER
TABLE_OWNER TABLE_NAME
TABLE_NAME COLUMN_NAME
UNIQUENESS COLUMN_POSITION
TABLESPACE_NAME COLUMN_LENGTH
LOGGING
STATUS
 Summary
In this lesson,you should have learned how to :

Create different types of indexes

Reorganize indexes
Drop Indexes
Get index information from the data dictionary
Reorganizing Data
 Objectives

Reorganize data using the Export and Import

utilities
Move data using transportable tablespaces
Moving Data Using Export and
Import

Export

Data Files
OS File

Import
Uses of Export and Import

Reorganize tables
Move data owned by one user to another user
Move data between databases :
- Development to production
- OLTP system to a data warehouse
Migrate the database to a different
- OS platform
- Release of the Oracle database
Repeat test runs during development or
upgrade
Perform a logical backup
 Export Modes

Table User Database

All objects All object in the

Table
owned by Database,except
Indexes
user,except objects owned by
Triggers
indexes and SYS
Constraints
triggers on
Grants
tables owned by
Analyze Method other users
Conventional and Direct Path
Export
Direct
Dump Conventional
File Export

Evaluating SQL Command

Buffer Processing
Private buffer Buffer Cache
or buffer cache Management
Read database
block

Database
 Using Export

Exp scott/tiger table=(dept,emp) \

> file=emp.dmp log=exp.log \
>compress=n direct=y

emp.dmp
export
DEPT and
EMP tables
Exp.log
 Using Import

Imp scott/tiger tables={dept,emp} \

> file=emp.dmp log=imp.log ignore=y

Imp.log
emp.dmp Import

Database
 Import Behavior

Order of Import :
Table -> Data -> B-Tree indexes -> constraints,
triggers, bitmap indexes
Tablespace used for the object :
- Same tablespace as in the source database,

if possible
- Users default tablespace
 Export and Import Guidelines

Use a parameter file to specify commonly used

command line options
Use CONSISTENT=Y only if exporting a small volume
of data
Do not use COMPRESS=Y if there are many deleted
rows
Improve performance by :
- Allocating Large buffer size
- Using direct path if using 7.3.3 or higher
 NLS Considerations in Export
and Import

Export Export Client

Character Set

Database
Character Set Import
Import client
Character Set

Possible character set conversion

 Data Transportation :
Transportable Tablespaces
OLTP Data mart

Datawarehouse

staging

Information
Distribution
Transporting Tablespaces

1. Make tablespace read only

2. Export metadata from source
3. Copy data files to target system
4. Transfer export file
5. Import metadata into target
6. If necessary, after the tablespace to read-write
Exporting and Importing
metadata
Exp sys/ FILE= 980501.dmp
TRANSPORT_TABLESPACE=y
TABLESPACES=sales_ts
TRIGGER=N CONSTRAINTS=N

Imp sys/ FILE=980501.dmp

TRANSPORT_TABLESPACE=y
DATAFILES=(..\disk1\sales01.dbf,
..\disk2\sales02.dbf)
 Transporting a Tablespace

System 1 System 2

/disk1
/u/d1 s980501.dmp
/u/d2 /disk2

SALES_TS SALES_TS
/u/d1/sales1.dbf \disk1\sales01.dbf

/u/d2/sales2.dbf \disk2\sales02.dbf

Read Only
Transportable Tablespaces :
Uses

Moves entire tablespace data

supports media recovery
source and target databases must :
- Be on the same operating system
- Run Oracle8i ,release 8.1 ,or above
- Have the same block size
- use the same character set.