Oracle Internal & Oracle Academy Use Only

Practices for Lesson 2:

Backup and Recovery
Configuration
Practices for Lesson 2: Overview

Overview
In these practices, you learn how to configure your database to enable recovery from various
losses. You verify the control file configuration, the fast recovery area (FRA), redo log groups,
ARCHIVELOG mode, and redundant archive log destinations.
How to configure your database for recovery:
 Ensure redundancy of control files. If a control file is damaged or lost, recovery is
easier if you have another copy.
 Review the fast recovery area configuration.
 Ensure that there are at least two redo log members in each group. If a redo log
member is damaged or lost, recovery is easier when you have an additional member in

Oracle Internal & Oracle Academy Use Only

the group.
 Place your database in ARCHIVELOG mode. In all cases, you will be able to recover
the database either completely or incompletely depending on which database files
have been damaged or lost.
 Configure redundant archive log destinations. In cases where you lost archive log files
and you need them to recover the database, you will be able to perform an incomplete
recovery, unless you have a duplicate version of the archive log in another destination.

Copyright © 2022, Oracle and/or its affiliates.

12 Practices for Lesson 2: Backup and Recovery Configuration

Practice 2-1: Verifying that the Control File is Multiplexed

Overview
In this practice, you verify that the control file is multiplexed.
A control file is a small binary file that describes the structure of the database. It must be
available for writing by the Oracle server whenever the database is mounted or opened. Without
this file, the database cannot be mounted, and recovery or re-creation of the control file is
required. Your database should have a minimum of two control files on different storage devices
to minimize the impact of a loss of one control file. The loss of a single control file causes the
instance to fail because all control files must be available at all times. However, recovery can be
a simple matter of copying one of the other control files. The loss of all control files is slightly
more difficult to recover from, but is not usually catastrophic.

Oracle Internal & Oracle Academy Use Only

Time Estimate
It should take approximately 15 minutes to complete this practice.

Tasks
1. Open a terminal window and start SQL*Plus and connect to the CDB root as the SYS user
with the SYSDBA privilege.
$ sqlplus / AS SYSDBA
...
SQL>
2. Find out how many control files exist in the database. The query returns the names of two
control files (control01.ctl and control02.ctl), which verifies that the control files
are multiplexed.
SQL> SELECT name FROM v$controlfile;

NAME
----------------------------------------------------------------
/u01/app/oracle/oradata/ORCLCDB/control01.ctl
/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl
SQL>
When the CDB was created, DBCA created two control files. When you use the CREATE
DATABASE command in SQL*Plus to create a database, you configure the
CONTROL_FILES parameter to generate two control files and set their names.

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 13

3. View the CONTROL_FILES parameter. Notice that the paths to the control files are stored in
this parameter. The results below are formatted for easier viewing.
SQL> SHOW PARAMETER control_files

NAME TYPE
---------------------------------------------------- -------
VALUE
--------------------------------------------------------
control_files string
/u01/app/oracle/oradata/ORCLCDB/control01.ctl,
/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl
SQL>

Oracle Internal & Oracle Academy Use Only

4. Create a parameter file (PFILE) from the server parameter file (SPFILE).
SQL> CREATE PFILE FROM SPFILE;

File created.

SQL>
5. Shut down the database instance in IMMEDIATE mode.
SQL> SHUTDOWN IMMEDIATE
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL>
6. Exit SQL*Plus.
SQL> EXIT
7. Create a directory for the new control file.
$ mkdir -p /u01/app/oracle/controlfiles_dir/ORCLCDB
$
8. Before you edit your PFILE, make a backup of it.
$ cp $ORACLE_HOME/dbs/initorclcdb.ora
$ORACLE_HOME/dbs/backup_initorclcdb.ora
$
9. Copy one of the control files to the directory you created in a previous step
(/u01/app/oracle/controlfiles_dir/ORCLCDB) and name the file
control03.ctl.
$ cp /u01/app/oracle/oradata/ORCLCDB/control01.ctl
/u01/app/oracle/controlfiles_dir/ORCLCDB/control03.ctl
$

Copyright © 2022, Oracle and/or its affiliates.

14 Practices for Lesson 2: Backup and Recovery Configuration

10. Open the PFILE (initorclcdb.ora) in the editor of your choice (vi or gedit) and add
the name of the new control file to the end of the list of control files. Include the path. Be
certain not to enter spaces between the single quotes and commas in the
control_files= line. Be certain that this line is one continuous line, without line breaks.
Save and close the file.
$ vi $ORACLE_HOME/dbs/initorclcdb.ora
...
*.control_files='/u01/app/oracle/oradata/ORCLCDB/control01.ctl',
'/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl',
'/u01/app/oracle/controlfiles_dir/ORCLCDB/control03.ctl'
...
$
11. Start SQL*Plus and connect to the root container as the SYS user with the SYSDBA privilege.

Oracle Internal & Oracle Academy Use Only

You are connected to an idle instance.
$ sqlplus / AS SYSDBA
…
Connected to an idle instance.

SQL>
12. Start the database instance.
SQL> STARTUP
ORACLE instance started.
…
Database opened.
SQL>
13. View the CONTROL_FILES parameter again.
SQL> SHOW PARAMETER control_files

NAME TYPE
------------------------------------ -----------
VALUE
------------------------------
control_files string
/u01/app/oracle/oradata/ORCLCDB/control01.ctl,
/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl
SQL>
Question: Why does the CONTROL_FILES parameter still show only two control files?
Answer: By default, the database instance starts up with the SPFILE. If an SPFILE does
not exist, then the instance starts up with a PFILE. In this case, both an SPFILE and PFILE
are present, so the SPFILE takes precedence. You configured the PFILE, not the SPFILE.
The SPFILE still contains only two references.

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 15

14. Re-create the third control file because the current version is no longer an exact copy of the
others.
a. Shut down the database instance with the IMMEDIATE option.
SQL> SHUTDOWN IMMEDIATE
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL>
b. Exit SQL*Plus.
SQL> EXIT
c. Use the cp command to re-create control03.ctl.

Oracle Internal & Oracle Academy Use Only

$ cp /u01/app/oracle/oradata/ORCLCDB/control01.ctl
/u01/app/oracle/controlfiles_dir/ORCLCDB/control03.ctl
$
15. Re-create the SPFILE from the updated PFILE.
a. Start SQL*Plus and connect to the CDB root as the SYS user with the SYSDBA privilege.
You are connected to an idle instance.
$ sqlplus / AS SYSDBA
…
Connected to an idle instance.

SQL>
b. Create the SPFILE.
SQL> CREATE SPFILE FROM PFILE;

File created.

SQL>
16. Start the database instance.
SQL> STARTUP
ORACLE instance started.
…
Database mounted.
Database opened.
SQL>
17. View the CONTROL_FILES parameter again. The third control file is now included in the list,
which indicates that the SPFILE is configured properly. The results below are formatted for
easier viewing.
SQL> SHOW PARAMETER control_files

Copyright © 2022, Oracle and/or its affiliates.

16 Practices for Lesson 2: Backup and Recovery Configuration

 NAME TYPE
------------------------------------ -----------
VALUE
------------------------------
control_files string
/u01/app/oracle/oradata/ORCLCDB/control01.ctl,
/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl,
/u01/app/oracle/controlfiles_dir/ORCLCDB/control03.ctl
SQL>
18. Query the V$CONTROLFILE view to confirm the number of control files. The result indicates
that three control files are defined. Exit out of SQL*Plus.
SQL> SELECT name FROM v$controlfile;

Oracle Internal & Oracle Academy Use Only

NAME
----------------------------------------------------------------
/u01/app/oracle/oradata/ORCLCDB/control01.ctl
/u01/app/oracle/fast_recovery_area/ORCLCDB/control02.ctl
/u01/app/oracle/controlfiles_dir/ORCLCDB/control03.ctl

SQL> exit
...
$
19. Run the cleanup02-01.sh script to clean up the environment.
$ /home/oracle/labs/admin/cleanup02-01.sh
...
SQL>
System altered.

SQL> Database closed.

Database dismounted.
ORACLE instance shut down.
SQL> ORACLE instance started.

Total System Global Area 2013264224 bytes

Fixed Size 9136480 bytes
Variable Size 486539264 bytes
Database Buffers 1509949440 bytes
Redo Buffers 7639040 bytes
Database mounted.
Database opened.
SQL>
NAME TYPE VALUE

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 17

 ------------------------ ----------- ---------------------------
control_files string /u01/app/oracle/oradata/ORCLCD
B/control01.ctl, /u01/app/orac
le/fast_recovery_area/ORCLCDB/
control02.ctl
SQL> Disconnected from Oracle Database 19c Enterprise Edition
Release 19.0.0.0.0 - Production
Version 19.3.0.0.0
$

Oracle Internal & Oracle Academy Use Only

Copyright © 2022, Oracle and/or its affiliates.

18 Practices for Lesson 2: Backup and Recovery Configuration

Practice 2-2: Configuring the Size of the Fast Recovery Area

Overview
In this practice, you review the fast recovery area (FRA) configuration and change its size to 12
GB.

Time Estimate
It should take approximately 5 minutes to complete this practice.

Tasks
1. Evaluate the space needed for the FRA. The amount of disk space to allocate for the FRA
depends on the size and activity levels of your database. As a general rule, the larger the

Oracle Internal & Oracle Academy Use Only

FRA, the more useful it is. Ideally, the FRA should be large enough for copies of your data
and control files, as well as for flashback, online redo, and archived logs needed to recover
the database with the backups kept based on the retention policy (covered in one of the
next practices). In short, the FRA should be at least twice the size of the database so that it
can hold one backup and several archived logs.
View the values of the DB_RECOVERY_FILE_DEST and DB_RECOVERY_FILE_
DEST_SIZE initialization parameters.
$ sqlplus / AS SYSDBA
...
SQL> SHOW PARAMETER db_recovery_file_dest

NAME TYPE
------------------------------------ -----------
VALUE
------------------------------
db_recovery_file_dest string
/u01/app/oracle/fast_recovery_area
db_recovery_file_dest_size big integer
14970M
SQL>
Question: Is the fast recovery area enabled?
Answer: Yes. The DB_RECOVERY_FILE_DEST and DB_RECOVERY_FILE_DEST_SIZE
parameters values are not null, indicating that the fast recovery area is enabled.

Question: What changes can you make to the fast recovery area?
Answer: You can change the location and size of the fast recovery area.

Question: Does changing the size of the fast recovery area require the database to be
restarted?

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 19

 Answer: No, a restart is not required for this change because the
DB_RECOVERY_FILE_DEST_SIZE parameter is dynamic.
2. Change the size of the fast recovery area to 12GB and set the scope to BOTH.
SQL> ALTER SYSTEM SET db_recovery_file_dest_size = 12G
SCOPE=both;

System altered.

SQL>
Note: If the archived redo log file destination fills up or cannot be written to, the database
will halt. You would then need to remove archived redo log files from the archived redo log
file destination so that the database could resume operations. This activity is covered in one

Oracle Internal & Oracle Academy Use Only

of the next practices.
3. View the DB_RECOVERY_FILE_DEST_SIZE initialization parameter again. The result
verifies that the size has been set to 12GB.
SQL> SHOW PARAMETER db_recovery_file_dest_size

NAME TYPE VALUE

-------------------------------- ----------- ------
db_recovery_file_dest_size big integer 12G
SQL>

Copyright © 2022, Oracle and/or its affiliates.

20 Practices for Lesson 2: Backup and Recovery Configuration

Practice 2-3: Verifying that the Redo Log File Is Multiplexed

Overview
Ensure that there are at least two redo log members in each group. If you are using file system
storage, then each member should be distributed on separate disks or controllers so that no
single equipment failure impacts an entire log group. The loss of an entire current log group is
one of the most serious media failures because it can result in data loss. The loss of a single
member of a multi-member log group does not affect database operation (other than causing an
alert to be published in the alert log). One set of members should be stored in the FRA.

Time Estimate
It should take approximately 20 minutes to complete this practice.

Oracle Internal & Oracle Academy Use Only

Tasks
1. Query V$LOGFILE to determine the configuration (number of members) for each redo log
group. The result shows that there are currently three log groups (1, 2, and 3) and only one
member in each group.
SQL> SELECT group#, status, member FROM v$logfile;

GROUP# STATUS MEMBER

------- ------- ----------------------------------------
3 /u01/app/oracle/oradata/ORCLCDB/redo03.log
2 /u01/app/oracle/oradata/ORCLCDB/redo02.log
1 /u01/app/oracle/oradata/ORCLCDB/redo01.log

SQL>
Question: Why is it recommended to have three groups when two would be sufficient?
Answer: The Oracle Database server treats the online redo log groups as a circular buffer
in which to store transaction information, filling one group and then moving on to the next.
After all groups have been written to, the Oracle Database server begins overwriting
information in the first log group. If the database is configured in ARCHIVELOG mode, the
LGWR cannot overwrite data in the first log group if it has not been archived.
Question: Can multiplexing redo logs impact database performance?
Answer: Multiplexing redo logs may heavily influence database performance because a
commit cannot complete until the transaction information has been written to the logs by
LGWR. You must place your redo log files on your fastest disks served by your fastest
controllers. If possible, do not place any other database files on the same disks as your
redo log files. Because only one group is written to at a given time, there is no performance
impact in having members from several groups on the same disk.
2. Add another member to each redo log group. Name each member redonnb.log, where
nn represents the group number.

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 21

 SQL> ALTER DATABASE ADD LOGFILE MEMBER
'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo01b.log' TO
GROUP 1;

Database altered.

SQL> ALTER DATABASE ADD LOGFILE MEMBER

'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo02b.log' TO
GROUP 2;

Database altered.

SQL> ALTER DATABASE ADD LOGFILE MEMBER

'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo03b.log' TO

Oracle Internal & Oracle Academy Use Only

GROUP 3;

Database altered.

SQL>
3. Verify that the redo log files are now multiplexed. The query result shows that each group
has two members, and therefore, the redo log files are multiplexed. Observe the INVALID
status of the newly added redo log members. This status is expected because the new
members have not yet been written to by LGWR. When a log switch occurs and the group
containing the new member becomes CURRENT, the new member's status will change to
null.
SQL> SELECT group#, status, member FROM v$logfile ORDER BY 1, 3;

GROUP# STATUS MEMBER

------- ------- ------------------------------------------------
1 INVALID
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo01b.log

1
/u01/app/oracle/oradata/ORCLCDB/redo01.log

2 INVALID
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo02b.log

2
/u01/app/oracle/oradata/ORCLCDB/redo02.log

3 INVALID
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo03b.log

Copyright © 2022, Oracle and/or its affiliates.

22 Practices for Lesson 2: Backup and Recovery Configuration

 3
/u01/app/oracle/oradata/ORCLCDB/redo03.log

6 rows selected.

SQL>
4. Switch the log files and observe the changes.
a. Find out which log group is the current log group. In this example, the query result
shows that group 1 is the current group. Your current group may be different.
SQL> SELECT group#, members, status FROM v$log;

Oracle Internal & Oracle Academy Use Only

GROUP# MEMBERS STATUS
---------- ---------- ----------------
1 2 CURRENT
2 2 INACTIVE
3 2 INACTIVE
SQL>
b. Switch the log files three times.
SQL> ALTER SYSTEM SWITCH LOGFILE;

System altered.

SQL> ALTER SYSTEM SWITCH LOGFILE;

System altered.

SQL> ALTER SYSTEM SWITCH LOGFILE;

System altered.

SQL>

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 23

 c. Query the V$LOGFILE view again. Notice that as a result of the log file switch, the new
members' statuses have changed to null.
SQL> SELECT group#, status, member FROM v$logfile ORDER BY 1, 3;

GROUP# STATUS
------ -------
MEMBER
----------------------------------------------------------------
1
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo01b.log

Oracle Internal & Oracle Academy Use Only

/u01/app/oracle/oradata/ORCLCDB/redo01.log

2
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo02b.log

2
/u01/app/oracle/oradata/ORCLCDB/redo02.log

3
/u01/app/oracle/fast_recovery_area/ORCLCDB/redo03b.log

3
/u01/app/oracle/oradata/ORCLCDB/redo03.log

6 rows selected.

SQL>
d. Query the V$LOG view again to learn which log group is now the current group. In this
example, the results show that the LGWR is writing to group 1. Your group may be
different. Your statuses may be different too. An INACTIVE status means the log group
is no longer needed for database instance recovery.
SQL> SELECT group#, members, status FROM v$log;

GROUP# MEMBERS STATUS

------- ---------- ----------------
1 2 CURRENT
2 2 INACTIVE
3 2 INACTIVE
SQL>

Copyright © 2022, Oracle and/or its affiliates.

24 Practices for Lesson 2: Backup and Recovery Configuration

 e. Switch the log file.
SQL> ALTER SYSTEM SWITCH LOGFILE;

System altered.

SQL>
f. Query the V$LOG view again. The current group has changed to group 2, and the
former current group's status is now ACTIVE. Your current group may be different. An
ACTIVE status means that the log group is active, but it’s not the current log group. It is
needed for crash recovery. It may be in use for block recovery.
SQL> SELECT group#, members, status FROM v$log;

Oracle Internal & Oracle Academy Use Only

GROUP# MEMBERS STATUS
------- ---------- ----------------
1 2 ACTIVE
2 2 CURRENT
3 2 INACTIVE
SQL>
g. Switch the log file again.
SQL> ALTER SYSTEM SWITCH LOGFILE;

System altered.

SQL>
h. Query the V$LOG view again. The current group has changed again to group 3, and the
status of both the other groups is now ACTIVE. Your current group may be different.
SQL> SELECT group#, members, status FROM v$log;

GROUP# MEMBERS STATUS

------- ---------- ----------------
1 2 ACTIVE
2 2 ACTIVE
3 2 CURRENT

SQL>
Question: Can the LGWR background process write to only one member of the CURRENT
group in case the other members are missing or damaged?
Answer: Yes, it can. As long as there is one member left in the CURRENT group, LGWR
can work.

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 25

5. To save space in your course practice environment, drop the redo log file members you
created in step 4.
a. Determine which redo log group is current. You cannot drop a member of the current
group.
SQL> SELECT group#, status FROM v$log;

GROUP# STATUS
---------- ----------------
1 ACTIVE
2 ACTIVE
3 CURRENT
SQL>

Oracle Internal & Oracle Academy Use Only

b. Drop the member in the previous group and then perform a log switch. In this example,
group 3 is current, so the command in this example drops a member in group 2.
SQL> ALTER DATABASE DROP LOGFILE MEMBER
'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo02b.log';

Database altered.

SQL> alter system switch logfile;

System altered.

SQL>
c. Drop the member in the next group and then perform a log switch.
SQL> ALTER DATABASE DROP LOGFILE MEMBER
'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo03b.log';

Database altered.

SQL> alter system switch logfile;

System altered.

SQL>

Copyright © 2022, Oracle and/or its affiliates.

26 Practices for Lesson 2: Backup and Recovery Configuration

 d. Drop the member in the final group and then perform a log switch.
SQL> ALTER DATABASE DROP LOGFILE MEMBER
'/u01/app/oracle/fast_recovery_area/ORCLCDB/redo01b.log';

Database altered.

SQL> alter system switch logfile;

System altered.

SQL>
e. Verify that each group now has only one member.

Oracle Internal & Oracle Academy Use Only

SQL> SELECT group#, members, status FROM v$log;

GROUP# MEMBERS STATUS

---------- ---------- ----------------
1 1 ACTIVE
2 1 ACTIVE
3 1 CURRENT

SQL>
f. Exit from SQL*Plus.
SQL> exit
…
$
g. Remove the physical files from the operating system.
$ rm /u01/app/oracle/fast_recovery_area/ORCLCDB/redo*.log
$
h. Verify that the redo log files have been removed.
$ ls /u01/app/oracle/fast_recovery_area/ORCLCDB
archivelog control02.ctl onlinelog
$

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 27

Practice 2-4: Configuring ARCHIVELOG Mode

Overview

In this practice, you configure your database for ARCHIVELOG mode so that redo logs are
archived.

Time Estimate
It should take approximately 10 minutes to complete this practice.

Tasks
1. Run the setup02-04.sh script to prepare the environment for this practice.

Oracle Internal & Oracle Academy Use Only

$ /home/oracle/labs/admin/setup02-04.sh
...
SQL> Database closed.
Database dismounted.
ORACLE instance shut down.
SQL> ORACLE instance started.

Total System Global Area 2013264224 bytes

Fixed Size 9136480 bytes
Variable Size 486539264 bytes
Database Buffers 1509949440 bytes
Redo Buffers 7639040 bytes
Database mounted.
SQL>
Database altered.

SQL>
Database altered.
...
$
2. Log in to SQL*Plus as the SYS user with the SYSDBA privilege.
$ sqlplus / AS SYSDBA
…
SQL>
3. Issue the ARCHIVE LOG LIST command to determine whether the database is in
ARCHIVELOG mode.
SQL> archive log list
Database log mode No Archive Mode
Automatic archival Disabled

Copyright © 2022, Oracle and/or its affiliates.

28 Practices for Lesson 2: Backup and Recovery Configuration

 Archive destination USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence 25
Current log sequence 27
SQL>
4. You must put the database in MOUNT mode to enable archiving. Shut down the database
instance, start it in MOUNT mode, and then enable archiving.
a. Shut down the database instance.
SQL> shutdown immediate
Database closed.
Database dismounted.
ORACLE instance shut down.
SQL>

Oracle Internal & Oracle Academy Use Only

b. Start the database instance and mount the database.
SQL> startup mount
ORACLE instance started.
…
Database mounted.
SQL>
c. Enable archiving.
SQL> alter database archivelog;

Database altered.

SQL>
d. Verify that the database is now in ARCHIVELOG mode.
SQL> archive log list
Database log mode Archive Mode
Automatic archival Enabled
Archive destination USE_DB_RECOVERY_FILE_DEST
Oldest online log sequence 25
Next log sequence to archive 27
Current log sequence 27
SQL>

e. Open the database.

SQL> alter database open;

Database altered.

SQL>

Copyright © 2022, Oracle and/or its affiliates.

Practices for Lesson 2: Backup and Recovery Configuration 29

5. Exit from SQL*Plus and close the terminal window.
SQL> EXIT
…
$

Oracle Internal & Oracle Academy Use Only

Copyright © 2022, Oracle and/or its affiliates.

30 Practices for Lesson 2: Backup and Recovery Configuration