Oracle RAC and APPS DBA

Creating Future

Cloning an Oracle Database

Posted by Narashim Reddy .R on July 27, 2009


Table of Contents
1. Purpose of the document
2. Document Usage
3. Overview of the Procedure
3.1 Introduction
3.2 Environments
3.3 Assumption
3.4 Cloning Process
3.4.1 Steps to be followed in Primary database:
3.4.2 Steps to be followed in the Secondary database server:
5. Conclusion

1. This document illustrates a process to clone an Oracle database using hot backup. The following sections are covered in this document

Overview
Environments
Assumptions
Cloning process

2. Document Usage

This document can be useful for any Oracle DBA for cloning an Oracle database. Generally cloning databases with the help of source database hot backup is a standard process to effectively and efficiently create a replica of the source database. The process illustrated in this document can be used by any DBA who wants to create a replica of an Oracle database with the help of hot backup and the associated archive logs.

3. Overview of the Procedure
3.1 Introduction

Cloning database is one of the routine DBA activities. Any DBA who administers various environments and is supporting the application development team for database activities has to periodically synchronize the acceptance/testing/development with the Production data.

This frequency of synchronization depends upon the business and application development teams requirements. Generally this process can be done with the below mentioned procedures

1)EXPORT/IMPORT utilities
2)DATA PUMP
3)Cold Backup Restoration
4)Hot Backup Recovery

Each of the above mentioned process has got its own pros and cons. To state a few vital reasons why Hot backup recovery process is chosen widely

a)You can recover to the most recent time period.
b)Doesn’t require outage of the source database
c)Requires less time compared to the other process
d)Best method for large databases

3.2 Environments

1. Primary or the Source database system
2. Secondary or the Target database system

3.3 Assumption

Primary database operated in ARCHIVELOG mode.
Secondary database to be created by cloning in a different server
Primary and Secondary environment Operating system are same.

3.4 Cloning Process
3.4.1 Steps to be followed in Primary database:

1)Perform HOTBACKUP of all the data files

2)LSN number

Login as sys as sysdba

Sql> Alter system archive log current;
Sql> Archive log list

Note down the Current Log Sequence

3)Control file trace

In Sql> prompt issue the statement

ALTER DATABASE BACKUP CONTROLFILE TO TRACE;

You can find the trace file in the udump directory.

3.4.2 Steps to be followed in the Secondary database server:

1)If the secondary database is to be created in a different server
FTP the following to a temporary location in the secondary database server.

a)hot backup of all the primary database data files
b)all the archive log files generated in the primary database after you start hot backup
c)control file trace of primary database
d)pfile of the primary database

2)Edit the pfile as it would be for secondary database server locations

a)Control_files- new locations
b)instance_name – new name
c)background_dump_dest – new location
d)core_dump_dest- new location
e)user_dump_dest – new location
f)log_archive_dest- new location

Save the pfile as init_.ora in

$ORACLE_HOME/dbs directory if UNIX
$ORACLE_HOME\database if WINDOWS

Where SID is the instance name given above in the edited pfile

3) In the control file trace Use the “SET #2. RESTLOGS case” CREATE CONTROLFILE section (Since we don’t use the primary database redolog files, we use this option). In that section Copy from “CREATE CONTROLFILE” to “CHARACTER SET ” to another file.
Save the file as cr_control.sql

Edit the following section in cr_control.sql

a)Use SET instead of REUSE

Old: CREATE CONTROLFILE REUSE DATABASE <”PRIMARY_DBNAME” >RESETLOGS

NEW: CREATE CONTROLFILE SET DATABASE <”SECONDARY_DBNAME”> RESETLOGS

b)Edit the LOGFILE paths as per the secondary database environment
c)Edit the DATAFILE paths as per the secondary database environment
d)Save the cr_control.sql file

4) Move the ftp ed Hot backup DATAFILE from temporary location of specific location as specified in the control file cr_control.sql

5) Move the archived logs from temporary location to the location specified in log_archive_dest in pfile

6) If Windows,

Create an oracle service using ORADIM utility

ORADIM -NEW -SID sid | -SRVC service [-INTPWD password] [-MAXUSERS number] [-STARTMODE a|m] [-PFILE file] [-TIMEOUT secs]

In the CMD prompt

SET ORACLE_SID=

Sqlplus “/ as sysdba”

You will get a Connected to an idle instance message.

7) If UNIX,

In the $ prompt give

EXPORT ORACLE_SID=

Sqlplus “/ as sysdba”

You will get a Connected to an idle instance message.

8)In the SQL> Prompt give

STARTUP NOMOUNT

9) Run the cr_control.sql file in the SQL> prompt

If Windows

@/cr_control.sql

If Unix

@\cr_control.sql if in Unix

You should get a message like Control File created

9) Recover the database using backup control file. In the SQL> prompt give

RECOVER DATABASE USING BACKUP CONTROLFILE UNTIL CANCEL;

Note: this can also be done with UNTIL TIME or UNTIL SCN provided you know the time or the SCN until which you want to perform the recovery

10)The recovery process will keep prompting for archive logs. Give return for each prompt of archive log.

Since the required archive logs from Primary database are restored in archive_log_dest of secondary, it will get applied for every return you give in the prompt.

Once the recovery process prompts for the archive log with the LSN number noted in the Primary database,

in step 2) of Steps to be done in Primary database section

Give CANCEL in the recovery sql prompt

11)Now you must get a message like

Media recovery cancelled.

12) Now in the SQL prompt give.

ALTER DATABASE OPEN RESETLOGS;

You should get a message like.

Database opened.

12)Bounce the database

SHUTDOWN IMMEDIATE;
STARTUP;

13)Check for errors in bdump.

14)Check log switch.

15)Check redo log archival

16)Compare the users, objects, size with the Primary database

17)Take a complete cold back up and maintain that as generation 0 backup.

You have cloned the database.

5. Conclusion
Hot backup recovery is one of the best methods to clone database because there is no outage or down time required in the primary database and data is replicated in secondary database to the most recent time period.

About these ads

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
Follow

Get every new post delivered to your Inbox.

Join 30 other followers

%d bloggers like this: