Skip Headers
Oracle® Database Backup and Recovery Reference
11g Release 2 (11.2)

Part Number E10643-01
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
View PDF

RESTORE

Purpose

Use the RESTORE command to restore, validate, or preview RMAN backups. Typically, you restore backups when a media failure has damaged a current datafile, control file, or archived redo log or before performing a point-in-time recovery.

Prerequisites

To restore datafiles to their current location, the database must be started, mounted, or open with the tablespaces or datafiles to be restored offline.

If you use RMAN in a Data Guard environment, then RMAN should be connected to a recovery catalog.

If you are performing a trial restore of the production database, then perform either of the following actions before restoring the database in the test environment:

If you do not perform either of the preceding actions, then RMAN assumes that you are restoring the production database and deletes flashback logs from the fast recovery area because they are considered unusable.

Usage Notes

The RESTORE command restores full backups, level 0 incremental backups, or image copies. You can restore files to their default location or a different location.

By default, RMAN examines read-only datafiles to make sure they exist, are readable, and have the correct checkpoint. If any of the conditions is not met, then RMAN restores the files. If all of the conditions are met, then RMAN does not restore the files.

Backup Selection

By default, RESTORE chooses the most recent backup set or file copy, that is, the file copy or backup set that needs the least media recovery. RMAN only restores backups created on the same type of channels allocated by the RESTORE command. For example, if you made backups of a datafile with DISK and sbt channels, and if only a DISK channel is allocated for the RESTORE command, then RMAN will not restore the sbt backups. If you do not manually allocate channels, then RMAN allocates all automatic channels that it possibly needs, subject to any restrictions imposed by the DEVICE TYPE option.

In an Oracle RAC configuration, RMAN automatically restores backups, control file copies, and datafile copies from channels that can read the files on tape or a local file system. For example, if channel ch1 connected to inst1 can read log 1000 from its tape drive, but channel ch2 connected to inst2 cannot read the same log from its tape drive, then ch1 cannot participate in restoring the log and so ch2 restores the log. Autolocation is automatically enabled when the channels have different PARMS or CONNECT settings.

If datafile names are symbolic links, then the control file stores the filenames of the link files but RMAN performs I/O on the datafiles pointed to by the link files. If a link file is lost and you restore a datafile without re-creating the symbolic link, then RMAN restores the datafile to the location of the link file rather than to the location pointed to by the link file.

See Also:

Oracle Database Backup and Recovery User's Guide for details on restore failover

Restore Operations Using Encrypted Backup Sets

As explained in "Encryption of Backup Sets", how RMAN handles encrypted backup sets during restore operations depends on the encryption mode with which the backup was created. You can use CONFIGURE and SET to manage the RMAN backup encryption settings for your database. Note the following restore considerations:

Note:

Wallet-based encryption is more secure than password-based encryption because no passwords are involved. You should use password-based encryption only when absolutely necessary because your backups need to be transportable.

Restore Failover

If a backup piece, image copy or proxy copy is inaccessible or if a block is corrupted, then RMAN performs restore failover. The RESTORE command automatically looks for another usable copy of a backup or image copy on the same device and other devices. If no usable copies are available, then RMAN searches for previous backups. RMAN continuously searches for previous usable backups until it has exhausted all possibilities. RMAN automatically uses eligible backups from previous database incarnations if required.

If you are restoring a datafile for which no backups are available, then RMAN will create an empty datafile with the checkpoint change as creation SCN. During recovery, all archived redo log files back to the creation of the datafile will be restored, and all changes during the history of the datafile will be reapplied to re-create its contents.

Location of Restored Datafiles

If you restore datafiles to the default location, then RMAN overwrites files with the same filenames. By default, RMAN will not restore a datafile if the datafile is in the correct place and its header contains the expected data. Note that RMAN does not scan the datafile body for corrupt blocks.

If RMAN detects that the default filename cannot be used (for example, the file may be an Oracle-managed file or on an Automatic Storage Management disk group), then RMAN attempts to create a new file in the same location or disk group.

To restore files to a nondefault location, use SET NEWNAME commands to rename the restored files and then use a SWITCH command to make the restored files current (as illustrated in Example 3-19). If you do not issue SWITCH commands, then RMAN considers the restored files as valid copies for use in future restore operations. Table 3-8 describes the behavior of the RESTORE, SET NEWNAME, and SWITCH commands.

Table 3-8 SET NEWNAME, SWITCH, and RESTORE

SET NEWNAME Run SWITCH Run RESTORE Behavior

No

N/A

RMAN restores the files to their current path names.

Yes

Yes

RMAN restores the files to the path names specified by SET NEWNAME. RMAN replaces the current datafile names in the control file with the names of the restored files. RMAN records the datafiles with the old names as datafile copies.

Yes

No

RMAN restores the files to the path names specified by SET NEWNAME. RMAN does not update the current datafile names in the control file. The restored files are listed in the RMAN repository as datafile copies.


Because tempfiles cannot be backed up and because no redo is ever generated for them, RMAN never restores or recovers tempfiles. RMAN does track the names of tempfiles, but only so that it can automatically re-create them when needed.

RMAN Behavior When Restoring Control Files

The behavior of RMAN when restoring control files depend on a variety of factors, which are summarized in Table 3-9. Required commands and options for restoring autobackups are summarized in Table 3-10.

Table 3-9 RESTORE CONTROLFILE Scenarios

RMAN Connection RESTORE CONTROLFILE; RESTORE CONTROLFILE FROM AUTOBACKUP; RESTORE CONTROLFILE ... TO 'filename'; RESTORE CONTROLFILE ... FROM 'media_handle' or TAG 'user_tag';

No catalog, target database started in NOMOUNT state

Error. Must specify FROM AUTOBACKUP.

Restores to CONTROL_FILES locations. See Table 3-10 for required commands and options.

Must specify FROM AUTOBACKUP. Restores only to filename.

First run SET DBID. Restores from specified file (cannot restore from TAG). If TO 'filename' not used, restores to all CONTROL_FILES locations.

No catalog, target database mounted or open

Error. Must use TO 'filename', where filename is not in CONTROL_FILES list.

Error. Must use TO 'filename', where filename is not in CONTROL_FILES list.

Restores only to filename, where filename is not in CONTROL_FILES list.

RMAN issues error RMAN-06496. Use TO 'filename' instead.

Catalog, target database started in NOMOUNT state

Restores to CONTROL_FILES locations. Run SET DBID only if DB_NAME not unique in catalog.

Only use with recovery catalog for testing.

Restores only to filename, where filename is not in CONTROL_FILES list.

Restores from specified file. If TO 'filename' not used, restores to all CONTROL_FILES locations.

Catalog, target database mounted or open

Error. Must use TO 'filename', where filename is not in CONTROL_FILES list.

Do not use with recovery catalog.

Restores only to filename, where filename is not in CONTROL_FILES list.

RMAN issues error RMAN-06496. Use TO 'filename' instead.


If you use RMAN in a Data Guard environment, then RMAN transparently converts primary control files to standby control files and vice versa. RMAN automatically updates filenames for datafiles, online redo logs, standby redo logs, and temp files when you issue RESTORE and RECOVER. The recovery catalog always contains the correct information about the backup filenames for each database, as explained in "RMAN Backups in a Data Guard Environment".

Control File and Server Parameter File Autobackup Options

When restoring an autobackup, the commands and options that you use depend on the autobackup type (control file or server parameter file) and location (inside or outside fast recovery area). The options are summarized in Table 3-10.

Table 3-10 RESTORE ... FROM AUTOBACKUP

Restore Object Autobackup Location Run SET DBID? Specify RECOVERY AREA on RESTORE? Specify DB_NAME or DB_UNIQUE_NAME on RESTORE? Run SET CONTROLFILE AUTOBACKUP FORMAT?

SPFILE

Recovery area

No

Yes

Yes

No

SPFILE

Outside recovery area

Yes

No

No

Only if autobackup is not in default location

Control file

Recovery area

No

Only if autobackup is in noncurrent recovery area

Only if autobackup is in noncurrent recovery area and uses a noncurrent DB_UNIQUE_NAME

No

Control file

Outside recovery area

Yes

No

No

Only if autobackup is not in default location


Syntax

restore::=

Description of restore.gif follows
Description of the illustration restore.gif

(restoreObject::=, restoreSpecOperand::=, deviceSpecifier::=, untilClause::=)

restoreObject::=

Description of restoreobject.gif follows
Description of the illustration restoreobject.gif

(archivelogRecordSpecifier::=, datafileSpec::=)

restoreSpecOperand::=

Description of restorespecoperand.gif follows
Description of the illustration restorespecoperand.gif

autoBackupOptList::=

Description of autobackupoptlist.gif follows
Description of the illustration autobackupoptlist.gif

Semantics

restore

This clause enables you to select which files you want to restore and specify parameters that control the behavior of the restore operation.

Syntax Element Description
restoreObject Specifies the files to be restored.
restoreSpecOperand Specifies options for the restoreObject clause.
CHANNEL channel_id Refer to the restoreSpecOperand clause.
CHECK LOGICAL Tests data and index blocks that pass physical corruption checks for logical corruption, for example, corruption of a row piece or index entry. If RMAN finds logical corruption, then it logs the block in the alert log and server session trace file.

If the total number of physical and logical corruptions detected in a file is less than its SET MAXCORRUPT setting, then the RMAN command completes and the database populates the V$DATABASE_BLOCK_CORRUPTION view with corrupt block ranges. If MAXCORRUPT is exceeded, then the command terminates without populating the views.

When restoring a backup datafile, RMAN honors the DB_BLOCK_CHECKSUM initialization parameter setting. RMAN clears the checksum if DB_BLOCK_CHECKSUM is set to false. If set to typical, then RMAN verifies the checksum when restoring from the backup and writing to the datafile. If the initialization parameter DB_BLOCK_CHECKSUM=typical, and if MAXCORRUPT is not set, then specifying CHECK LOGICAL detects all types of corruption that are possible to detect.

Note: The MAXCORRUPT setting represents the total number of physical and logical corruptions permitted on a file.

DEVICE TYPE deviceSpecifier Allocates automatic channels for the specified device type only. For example, if you configure automatic disk and tape channels, and issue RESTORE ... DEVICE TYPE DISK, then RMAN allocates only disk channels. You must have already configured a device type by using CONFIGURE (except for DISK, which is preconfigured) before specifying the DEVICE TYPE option.

Note: You cannot manually allocate channels within a RUN block and then run RESTORE with the DEVICE TYPE clause.

See Also: deviceSpecifier

FORCE Overrides the restartable restore feature and restores all files regardless of whether they need to be restored. If you do not specify FORCE, then RMAN restores a file only if its header information does not match the information in the control file.
FROM BACKUPSET Specifies that RMAN should restore from backup sets only. By default RESTORE chooses the file copy or backup set that needs the least media recovery.

If you use the FROM BACKUPSET option, then channels for the appropriate type of storage devices must be allocated for the backup sets that need to be restored. For example, if needed backups are only available on tape, and no sbt channels have been allocated, then RMAN cannot find a candidate backup set to restore, and the RESTORE command fails.

FROM DATAFILECOPY Specifies that RMAN should restore datafile copies only. By default RESTORE chooses the file copy or backup set that needs the least media recovery. If you use the FROM DATAFILECOPY option, then the allocated channels must be of DEVICE TYPE DISK.
FROM TAG tag_name Refer to the restoreSpecOperand clause.
PREVIEW Reports—but does not restore—the backups and archived redo log files that RMAN could use to restore and recover the database to the specified time. RMAN queries the metadata and does not actually read the backup files.

The RESTORE ... PREVIEW output is in the same format as the LIST BACKUP output (see Example 3-24).

Some media managers provide status information to RMAN about which backups are offsite. Offsite backups are stored in a remote location, such as a secure storage facility, and cannot be used without retrieving media.

Offsite backups are marked as AVAILABLE in the RMAN repository even though the media must be retrieved from storage before the backup can be restored. If RMAN attempts to restore a offsite backup, then the restore operation fails. RESTORE ... PREVIEW can identify backups needed for a RESTORE operation that are stored on media that requires retrieval. The output indicates whether backups are stored offsite.

If a needed backup is stored offsite, but the media manager does not support offsite backups, then your options are:

  • Use CHANGE ... UNAVAILABLE to prevent RMAN from selecting the needed offsite backups, and attempt the RESTORE ... PREVIEW operation again to determine whether RMAN selects another offsite backup. When RMAN does not select any offsite backups, you can perform the restore operation.

  • Use RESTORE ... PREVIEW with the RECALL option.

See Also: LIST, specifically the BACKUPS and SUMMARY options, and the RECOVER ... VALIDATE HEADER command

   RECALL Instructs the media manager to retrieve the backup media needed for the specified restore operation from offsite storage (see Example 3-25).

Note: This option only works if your media manager supports this functionality. You can use RESTORE ... PREVIEW periodically to monitor whether the needed backups are stored locally again.

   SUMMARY Summarizes the backups that RMAN would restore. The output is in the same format as the output of the LIST BACKUPS ... SUMMARY command.
SKIP READONLY Does not restore read-only files.

TO RESTORE POINT
restore_point_name
Specifies a restore point, with the SCN at which the restore point was created as the upper, inclusive limit. Because the limit is inclusive, RMAN selects only files that can be used to restore files up to and including the SCN corresponding to the restore point.
untilClause Limits the selection to backup sets or file copies that are suitable for a point-in-time recovery to the specified time, SCN, or log sequence number.

In the absence of any other criteria, RMAN selects the most current file copy or backup set to restore. Note that the time specified in the UNTIL clause must fall within the current database incarnation.

See Also: untilClause

VALIDATE Specifies that RMAN should decide which backup sets, datafile copies, and archived redo log files need to be restored, and then validate them (see Example 3-26). No files are restored.

For files on both disk and tape, RMAN reads all blocks in the backup piece or image copy. RMAN also validates offsite backups. The validation is identical to a real restore operation except that RMAN does not write output files.

Note: If you use RESTORE with the VALIDATE option, then the database can be open with datafiles online.

See Also: VALIDATE

   HEADER Reports and validates—but does not restore—the backups that RMAN could use to restore to the specified time.

When you specify this option, RMAN performs the same functions as when you run RESTORE with the PREVIEW option. However, in addition to listing the files needed for restore and recovery, RMAN validates the backup file headers to determine whether the files on disk or in the media management catalog correspond to the metadata in the RMAN repository.

See Also: The descriptions of the RESTORE PREVIEW option and the RECOVER ... VALIDATE HEADER option


restoreObject

This subclause specifies the objects to be restored: control files, datafiles, archived redo log files, or the server parameter file. Note that RMAN does not support backup and recovery of the change tracking file. RMAN re-creates the change tracking file after database restore and recovery; the next incremental backup after any recovery is able to use the file. Thus, restore and recovery has no user-visible effect on change tracking.

Syntax Element Description
archivelogRecordSpecifier Restores the specified range of archived redo log files.

The default restore location is DB_RECOVERY_FILE_DEST (if one of LOG_ARCHIVE_DEST_n is configured to USE_DB_RECOVERY_FILE_DEST either implicitly or explicitly). Otherwise, the default restore filenames are constructed with the LOG_ARCHIVE_FORMAT and LOG_ARCHIVE_DEST_1 initialization parameters of the target database. These parameters combine in a port-specific fashion to derive the name of the restored log. You can override the default location with the SET ARCHIVELOG DESTINATION command.

Because the RECOVER command automatically restores archived redo log files as needed, you should seldom need to restore logs manually. Possible reasons for manually restoring archived redo log files are to speed up recovery, to stage the logs to multiple destinations, or to analyze the log contents after a point-in-time recovery. To restore logs from a previous incarnation without shutting down the database, you can use RESTORE ARCHIVELOG with:

  • ... FROM SCN

  • ...SCN BETWEEN... AND

  • FROM SCN ... INCARNATION <integer>

  • FROM SCN... INCARNATION ALL

Note: The database can be started, mounted, or open for this operation.

See Also: archivelogRecordSpecifier.

CONTROLFILE Restores either a standby or backup control file depending on the target database role.

If the control file is lost, then restore the control file (see Table 3-9) and restore the database after mounting the restored control file. You must always run the RECOVER command after mounting a restored control file and you must open the database with the RESETLOGS option.

Note: If the target database is not mounted, and if RMAN is not connected to a recovery catalog, then you must specify the FROM AUTOBACKUP clause with RESTORE CONTROLFILE. If the autobackup is in a nondefault format, then first use the SET CONTROLFILE AUTOBACKUP FORMAT command to specify the format. If the target database is mounted or open, then you must specify the TO filename clause with RESTORE CONTROLFILE.

When you run RESTORE with a backup control file while connected to a recovery catalog (see Example 3-20), RMAN automatically updates the control file to reflect the structure of the restored database based on the metadata in the catalog.

   TO 'filename' Restores the control file to the specified filename.

Table 3-9 explains RMAN behavior when restoring the control file with the TO clause.

DATABASE Restores all datafiles in the database except those that are offline. By default, RMAN restores datafiles in read-only tablespaces.

Unlike BACKUP DATABASE, RESTORE DATABASE does not automatically include the control file and the server parameter file—you must issue additional RESTORE CONTROLFILE and RESTORE SPFILE commands to restore these files.

Note: To restore offline datafiles you must use RESTORE DATAFILE or RESTORE TABLESPACE.


   SKIP [FOREVER]
   TABLESPACE
   tablespace_name
Excludes the specifies tablespaces from the restore operation. This option is useful to avoid restoring tablespaces containing temporary data.

Specifying the FOREVER keyword does not change the behavior of SKIP. The FOREVER keyword exists solely to maintain compatible syntax between RESTORE SKIP FOREVER and RECOVER SKIP FOREVER.

DATAFILE datafileSpec Restores the datafiles specified by filename or absolute datafile number (see Example 3-19).

Note: Do not specify a datafile more than once in a restore job. For example, the following command is invalid because datafile 1 is both specified explicitly and implied by the SYSTEM tablespace:

RESTORE TABLESPACE SYSTEM DATAFILE 1;

See Also: datafileSpec

PRIMARY CONTROLFILE Restores a control file for a primary database in a Data Guard environment.

RMAN restores either a normal or standby control file as appropriate, depending on the most recent database role known to the recovery catalog (RC_SITE.DATABASE_ROLE) for the target database. The purpose of this option to override the default setting in cases where the most recent database role is out-of-date.

Assume that you perform a switchover from primary database dgny to standby database dgsf, so that dgsf is the new primary database. You want to restore a control file on dgsf, but the recovery catalog was not resynchronized and still shows dgsf as a standby database. In this case, you can specify PRIMARY CONTROLFILE to override the default RMAN behavior and restore a normal control file.

SPFILE Restores a primary or standby server parameter file to the location from which it was backed up. RMAN cannot overwrite a server parameter file currently in use by the target database.

By default RMAN restores the most current server parameter file. Specify the UNTIL or TAG options to restore older versions of the server parameter file.

If the server parameter file is lost, then connect RMAN to the target database (and recovery catalog if used) and run SET DBID. Run STARTUP FORCE NOMOUNT before running RESTORE SPFILE. Then run STARTUP FORCE to restart the database instance with the restored server parameter file.

Note: If the target database is not mounted, and if RMAN is not connected to a recovery catalog, then you must specify the FROM AUTOBACKUP clause with RESTORE SPFILE. If the autobackup is in a nondefault format, then first use the SET CONTROLFILE AUTOBACKUP FORMAT command to specify the format. If the target database is started, mounted, or open, and if the database was started with a server parameter file, then you must specify the TO filename clause with RESTORE SPFILE.

   TO [PFILE] 'filename' Restores a primary or standby server parameter file to the location specified by the TO clause. Specify PFILE to save the server parameter file as a text-based initialization parameter file.

   FOR DB_UNIQUE_NAME
   db_unique_name
Specifies the DB_UNIQUE_NAME for the target database when the instance is not started. This parameter is only useful in a Data Guard environment.

When FOR DB_UNIQUE_NAME is specified, RMAN can locate the correct RMAN configurations for the host on which the SPFILE is being restored and use them to access backup devices. Otherwise, RMAN cannot choose the correct channel configurations and returns an RMAN-6758 error.

In a Data Guard environment, the primary and standby hosts may have different channel configurations for communicating with their associated SBT backup and disk devices. If both the primary and standby databases are known to the recovery catalog, then the configuration settings for both databases are recorded in the recovery catalog. Because the two databases have the same DB_NAME, the records in the recovery catalog can only be distinguished with the DB_UNIQUE_NAME initialization parameter.

Note: Using RESTORE SPFILE when the DB_NAME is not unique in the recovery catalog produces an RMAN-6758 error.

See Also: Oracle Data Guard Concepts and Administration for a detailed procedure for restoring the server parameter file in a Data Guard environment

   TO 'filename' Restores the standby control file to the specified filename. Table 3-9 explains the RMAN behavior when restoring the control file with the TO clause.
STANDBY CONTROLFILE Restores a control file for a standby database. RMAN can transparently restore a normal control file backup and make it usable for a standby database.

RMAN restores either a normal or standby control file as appropriate, depending on the most recent database role known to the recovery catalog (RC_SITE.DATABASE_ROLE) for the target database. The purpose of this option to override the default setting in cases where the most recent database role is out-of-date. Assume that you perform a switchover from primary database dgny to standby database dgsf, so that dgsf is the new primary database. Later, you make dgny a standby database for dgsf. You want to restore a control file on dgny, but the recovery catalog was not resynchronized and still shows dgny as a primary database. In this case, you can specify STANDBY CONTROLFILE to override the default RMAN behavior and restore a standby control file.

If you restore the control file of a database whose DB_UNIQUE_NAME is known to the recovery catalog, then RMAN updates all filenames in the control file to filenames known to the recovery catalog. Any filenames explicitly renamed with ALTER DATABASE RENAME FILE take precedence over the filenames in the recovery catalog.

See Also: Table 3-9 for restrictions and usage notes

Note: You must always run the RECOVER command after mounting a restored control file, and must also always open the database with the RESETLOGS option.

TABLESPACE tablespace_name Restores all datafiles in the specified tablespaces (see Example 3-18).

RMAN translates the tablespace name internally into a list of datafiles. If you rename a tablespace (for example, from users to customers), then so long as an additional tablespace with the old name (users) has not been created, you can use either the old name (users) or the new name (customers) for the tablespace. RMAN detects that the tablespace has changed its name and updates the recovery catalog on the next resynchronization.

Note: RMAN can back up and restore dictionary-managed temporary tablespaces, but it cannot back up locally managed temporary tablespaces. However, RMAN automatically re-creates locally managed temporary tablespaces after restoring the database.


restoreSpecOperand

This subclause specifies options for the restoreObject clause. These parameters override the parameters with the same name at the RESTORE command level.

Syntax Element Description
CHANNEL channel_id Specifies the case-sensitive name of a channel to use for this restore operation. If you do not specify a channel, then RESTORE uses any available channel allocated with the correct device type.
FROM AUTOBACKUP Restores a control file autobackup (see Example 3-21).

This option is only valid on the RESTORE CONTROLFILE and RESTORE SPFILE commands. When restoring either type of file in NOCATALOG mode, the FROM AUTOBACKUP clause is required.

RMAN begins the search on the current day or on the day specified with the SET UNTIL. On the first day searched, the search begins with sequence number 256 (or the sequence number specified by MAXSEQ, if provided) and counts back to sequence 0. If no autobackup is found in the current or SET UNTIL day, then RMAN checks preceding days, starting with sequence 256 and counting back to 0. The search continues up to MAXDAYS days (default of 7, maximum of 366) prior to the current or SET UNTIL day. If no autobackup is found within MAXDAYS days, then RMAN signals an error and the command stops.

See Also: Table 3-9 for restrictions and usage notes.

   autoBackupOptList Specifies parameters that control the search for a control file autobackup.
   'media_handle' Specifies the name of the control file copy or backup piece containing a control file. The media_handle can be any backup piece that contains a backup of a control file: the control file backup does not need to be an autobackup.

See Also: Table 3-9 for restrictions and usage notes.

FROM TAG tag_name Overrides the default selection of the most recent backups or file copy available. The tag restricts the automatic selection to backup sets or file copies created with the specified tag. If multiple backup sets or copies have a matching tag, then RMAN selects the most recent one. Tag names are not case sensitive.

See Also: BACKUP for a description of how a tag can be applied to an individual copy of a duplexed backup set, and for a description of the default filename format for tags.


autoBackupOptList

This subclause specifies parameters that control the search for a control file autobackup.

Syntax Element Description
DB_NAME database_name Provides a DB_NAME to be used in searching for control file autobackups. See Table 3-10 to determine when to set this parameter.

The default value of the DB_UNIQUE_NAME initialization parameter is the DB_NAME initialization parameter setting. If no DB_UNIQUE_NAME initialization parameter is set for a target database, then use either RESTORE ... DB_NAME or RESTORE ... DB_UNIQUE_NAME. If the DB_UNIQUE_NAME initialization parameter setting for a target database is different from DB_NAME, then use RESTORE ... DB_UNIQUE_NAME.

MAXDAYS integer Limits the search for a control file autobackup to within the specified number of days in the past.
MAXSEQ integer Specifies the highest sequence number for the control file autobackup search.
RECOVERY AREA 'pathname' Specifies a path to the fast recovery area to search for autobackups. RECOVERY AREA and DB_RECOVERY_FILE_DEST are synonyms. See Table 3-10 to determine when to set this parameter.

DB_RECOVERY_FILE_DEST
'pathname'
RECOVERY AREA and DB_RECOVERY_FILE_DEST are synonyms.

   DB_NAME
   database_name
Provides a DB_NAME to be used in searching for control file autobackups. See Table 3-10 to determine when to set this parameter.

The default value of the DB_UNIQUE_NAME initialization parameter is the DB_NAME initialization parameter setting. If no DB_UNIQUE_NAME initialization parameter is set for a target database, then use either RESTORE ... DB_NAME or RESTORE ... DB_UNIQUE_NAME. If the DB_UNIQUE_NAME initialization parameter setting for a target database is different from DB_NAME, then use RESTORE ... DB_UNIQUE_NAME.


   DB_UNIQUE_NAME
   db_unique_name
Specifies the DB_UNIQUE_NAME of the database in the specified fast recovery area that is the target of the restore operation.

The default value of the DB_UNIQUE_NAME initialization parameter is the DB_NAME initialization parameter setting. If no DB_UNIQUE_NAME initialization parameter is set for a target database, then use either RESTORE ... DB_NAME or RESTORE ... DB_UNIQUE_NAME. If the DB_UNIQUE_NAME initialization parameter setting for a target database is different from DB_NAME, then use RESTORE ... DB_UNIQUE_NAME.


Examples

Example 3-18 Restoring a Tablespace

This example takes a tablespace offline, restores it, then performs media recovery.

SQL "ALTER TABLESPACE users OFFLINE IMMEDIATE"; 
RESTORE TABLESPACE users; 
RECOVER TABLESPACE users;  
SQL "ALTER TABLESPACE users ONLINE";

Example 3-19 Setting a New Name for a Restored Datafile

Assume that /disk1, which contains datafile 9, suffers a media failure. This example specifies a new name for the datafile, restores it, updates the control file to use the new name, recovers it, and then brings it online:

RUN
{
  SQL "ALTER DATABASE DATAFILE 9 OFFLINE";
  SET NEWNAME FOR DATAFILE 9 TO '/disk2/users01.dbf';
  RESTORE DATAFILE 9;
  SWITCH DATAFILE ALL;
  RECOVER DATAFILE 9;
  SQL "ALTER DATABASE DATAFILE 9 ONLINE";
}

Example 3-20 Restoring the Control File When Using a Recovery Catalog

Assume that you want to restore the control file backup with the tag monday_cf_backup. You start the RMAN client, connect to the target and recovery catalog databases, and run the following commands:

RUN
{ # SET DBID is not necessary when RMAN is connected to a recovery catalog
  STARTUP FORCE NOMOUNT;
  RESTORE CONTROLFILE FROM TAG 'monday_cf_backup';
  ALTER DATABASE MOUNT;
  RESTORE DATABASE;
  RECOVER DATABASE;
}
ALTER DATABASE OPEN RESETLOGS; # required after recovery with backup control file

RMAN restores the control file to its default location and replicates it automatically to all CONTROL_FILES locations. RMAN mounts the control file and restores and recovers the database. RMAN automatically updates the control file to reflect the structure of the restored database based on the metadata in the recovery catalog.

Example 3-21 Recovering the Database with a Control File Autobackup

Assume that the control file and some datafiles are lost and need to be restored from tape. Because RMAN does not use a recovery catalog in this scenario, the SET DBID command is necessary to identify the control file to be restored. The example restores the control file from tape, mounts the database, and then restores and recovers the database.

CONNECT TARGET /
STARTUP FORCE NOMOUNT;
SET DBID 36508508;  # required when restoring control file in NOCATALOG mode
RUN
{
  ALLOCATE CHANNEL c1 DEVICE TYPE sbt;
  RESTORE CONTROLFILE FROM AUTOBACKUP;
  ALTER DATABASE MOUNT;
  RESTORE DATABASE;
  RECOVER DATABASE;
}
ALTER DATABASE OPEN RESETLOGS;

Example 3-22 Restoring a Control File Autobackup to a Nondefault Location

This example is a variation on Example 3-21. In this scenario, the control file autobackup is located on disk in a nondefault location. RMAN starts searching for backups with a sequence number of 20, and searches backward for 5 months:

CONNECT TARGET /
STARTUP FORCE NOMOUNT
SET DBID 36508508;  # required when restoring control file in NOCATALOG mode
RUN
{
  SET CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK 
    TO '/disk1/prod_cf_auto_%F';
  RESTORE CONTROLFILE TO '/tmp/cf_auto.dbf' FROM AUTOBACKUP 
    MAXSEQ 20 MAXDAYS 150;
  ALTER DATABASE MOUNT;
  RESTORE DATABASE;
  RECOVER DATABASE;
}
ALTER DATABASE OPEN RESETLOGS;

Example 3-23 Restoring a Server Parameter File Autobackup to the Current Location

The following series of commands restores the current server parameter file in NOCATALOG mode and then starts the instance with the restored server parameter file.

CONNECT TARGET /
SET DBID 1620189241; # set dbid to dbid of target database
STARTUP FORCE NOMOUNT; # start instance with dummy SPFILE
RUN
{
  ALLOCATE CHANNEL c1 DEVICE TYPE sbt;
  RESTORE SPFILE FROM AUTOBACKUP; # FROM AUTOBACKUP needed in NOCATALOG mode
  STARTUP FORCE; # startup with restored SPFILE
}

Example 3-24 Previewing Backups

This example shows the results of a RESTORE ... PREVIEW command, which identifies the backup sets RMAN selects for use in restoring archived redo log files.

RMAN> RESTORE ARCHIVELOG ALL DEVICE TYPE sbt PREVIEW;
 
Starting restore at 01-MAR-07
released channel: ORA_SBT_TAPE_1
allocated channel: ORA_SBT_TAPE_1
channel ORA_SBT_TAPE_1: SID=85 device type=SBT_TAPE
channel ORA_SBT_TAPE_1: Oracle Secure Backup
 
List of Backup Sets
===================
 
BS Key  Size       Device Type Elapsed Time Completion Time
------- ---------- ----------- ------------ ---------------
53      1.25M      SBT_TAPE    00:00:18     01-MAR-07
        BP Key: 53   Status: AVAILABLE  Compressed: NO  Tag: TAG20070301T150155
        Handle: 2aibhej3_1_1   Media: RMAN-DEFAULT-000001
 
  List of Archived Logs in backup set 53
  Thrd Seq     Low SCN    Low Time  Next SCN   Next Time
  ---- ------- ---------- --------- ---------- ---------
  1    8       526376     01-MAR-07 527059     01-MAR-07
  1    9       527059     01-MAR-07 527074     01-MAR-07
  1    10      527074     01-MAR-07 527091     01-MAR-07
  1    11      527091     01-MAR-07 527568     01-MAR-07
  1    12      527568     01-MAR-07 527598     01-MAR-07
validation succeeded for backup piece
Finished restore at 01-MAR-07

Example 3-25 Recalling Offsite Backups from Offsite Storage

When used with a media manager that reports information about offsite storage of backups and supports recalling offsite backups, RESTORE ... PREVIEW RECALL requests that any media needed to restore archived redo log files from backup be recalled from offsite storage.

RMAN> RESTORE ARCHIVELOG ALL PREVIEW RECALL;

Starting restore at 10-JUN-06
using channel ORA_DISK_1
using channel ORA_SBT_TAPE_1
 
 
List of Backup Sets
===================
 
BS Key  Size       Device Type Elapsed Time Completion Time
------- ---------- ----------- ------------ ---------------
31      12.75M     SBT_TAPE    00:00:02     10-JUN-06     
        BP Key: 33   Status: AVAILABLE  Compressed: NO  Tag: TAG20050610T152755
        Handle: 15gmknbs   Media: /v1,15gmknbs
 
  List of Archived Logs in backup set 31
  Thrd Seq     Low SCN    Low Time  Next SCN   Next Time
  ---- ------- ---------- --------- ---------- ---------
  1    1       221154     06-JUN-06 222548     06-JUN-06
  1    2       222548     06-JUN-06 222554     06-JUN-06
  1    3       222554     06-JUN-06 222591     06-JUN-06
  1    4       222591     06-JUN-06 246629     07-JUN-06
  1    5       246629     07-JUN-06 262451     10-JUN-06
 
BS Key  Size       Device Type Elapsed Time Completion Time
------- ---------- ----------- ------------ ---------------
32      256.00K    SBT_TAPE    00:00:01     10-JUN-06     
        BP Key: 34   Status: AVAILABLE  Compressed: NO  Tag: TAG20050610T153105
        Handle: 17gmknhp_1_1   Media: /v1,17gmknhp_1_1
 
  List of Archived Logs in backup set 32
  Thrd Seq     Low SCN    Low Time  Next SCN   Next Time
  ---- ------- ---------- --------- ---------- ---------
  1    6       262451     10-JUN-06 262547     10-JUN-06
  1    7       262547     10-JUN-06 262565     10-JUN-06
 
Initiated recall for the following list of offsite backup files
==========================================================
        Handle: 15gmknbs   Media: /v1,15gmknbs
Finished restore at 10-JUN-06

Example 3-26 Validating the Restore of a Backup

The following example illustrates using RESTORE... VALIDATE to confirm that backups required to restore the database are present on disk or tape, readable, and not corrupted:

RMAN> RESTORE DATABASE VALIDATE;
 
Starting restore at 01-MAR-07
using channel ORA_DISK_1
allocated channel: ORA_SBT_TAPE_1
channel ORA_SBT_TAPE_1: SID=85 device type=SBT_TAPE
channel ORA_SBT_TAPE_1: Oracle Secure Backup
 
channel ORA_DISK_1: starting validation of datafile backup set
channel ORA_DISK_1: reading from backup piece /disk2/PROD/backupset/2007_03_01/o1_mf_nnndf_TAG20070301T161038_2ygtvzg0_.bkp
channel ORA_DISK_1: piece handle=/disk2/PROD/backupset/2007_03_01/o1_mf_nnndf_TAG20070301T161038_2ygtvzg0_.bkp tag=TAG20070301T161038
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: validation complete, elapsed time: 00:00:16
Finished restore at 01-MAR-07