3
RMAN Recovery Concepts

This chapter describes the basic concepts involved in using RMAN to restore, recover, and duplicate databases.

This chapter contains these topics:

• Restoring Files with RMAN
• Datafile Media Recovery with RMAN
• Block Media Recovery with RMAN
• Database Duplication with RMAN
• Physical Standby Database Creation with RMAN

Restoring Files with RMAN

Use the RMAN RESTORE command to restore the following types of files from disk or other media:

• Database (all datafiles)
• Tablespaces
• Control files
• Archived redo logs
• Server parameter files

Because a backup set is in a proprietary format, you cannot simply copy it as you would a backup database file created with an operating system utility; you must use the RMAN RESTORE command to extract its contents. In contrast, the database can use image copies created by the RMAN BACKUP AS COPY command without additional processing.

Mechanics of Restore Operations

RMAN automates the procedure for restoring files. You do not need to go into the operating system, locate the backup that you want to use, and manually copy files into the appropriate directories. When you issue a RESTORE command, RMAN directs a server session to restore the correct backups to either:

• The default location, overwriting the files with the same name currently there
• A new location, which you can specify with the SET NEWNAME command

To restore a datafile, either mount the database or keep it open and take the datafile to be restored offline. When RMAN performs a restore, it creates the restored files as datafile image copies and records them in the repository. The following table describes the behavior of the RESTORE, SET NEWNAME, and SWITCH commands.

Run SET NEWNAME?	RESTORE Behavior	Run SWITCH?
No	RMAN restores the files to their current path names and immediately removes the repository records for the datafile copies created during the restore.	N/A
Yes	RMAN restores the files to the path names specified by SET NEWNAME and does not remove the repository records for the datafile copies created during the restore.	If yes, then RMAN updates the datafile names in the control file to the names of the restored files; if no, then RMAN does not update the filenames in the control file and the restored files become datafile copies.

For example, if you restore datafile ?/oradata/trgt/tools01.dbf to its default location, then RMAN restores the file ?/oradata/trgt/tools01.dbf and overwrites any file that it finds with the same filename. If you run a SET NEWNAME command before you restore a file, then RMAN creates a datafile copy with the name that you specify. For example, assume that you run the following commands:

SET NEWNAME FOR DATAFILE '?/oradata/trgt/tools01.dbf' TO '/tmp/tools01.dbf';
RESTORE DATAFILE '?/oradata/trgt/tools01.dbf';

In this case, RMAN creates a datafile copy of ?/oradata/trgt/tools01.dbf named /tmp/tools01.dbf and records it in the repository. To change the name for datafile ?/oradata/trgt/tools01.dbf to /tmp/tools01.dbf in the control file, run a SWITCH command so that RMAN considers the restored file as the current database file. For example:

SWITCH DATAFILE '/tmp/tools01.dbf' TO DATAFILECOPY '?/oradata/trgt/tools01.dbf';

The SWITCH command is the RMAN equivalent of the SQL statement ALTER DATABASE RENAME FILE.

File Selection in Restore Operations

RMAN uses the repository to select the best available backups for use in the restore operation. The most recent backup available, or the most recent backup satisfying any UNTIL clause specified in the RESTORE command, is always the preferred choice. If two backups are from the same point in time, RMAN prefers to use image copies over backup sets because RMAN can restore more quickly from image copies than from backup sets (especially those stored on tape).

All specifications of the RESTORE command must be satisfied before RMAN restores a backup. Unless limited by the DEVICE TYPE clause, the RESTORE command searches for backups on all device types of configured channels.

If no available backup in the repository satisfies all the specified criteria, then RMAN returns an error during the compilation phase of the restore job. If you manually allocate channels, and if the file cannot be restored because no backups exist on the device types allocated in the job, then create a new job specifying channels for devices containing the existing backups. This problem does not occur when you configure automatic channels.

Restore Failover

During a RESTORE operation, if a backup piece, image copy or proxy copy is inaccessible (for instance, deleted from the device) or a block in the backup is corrupted, then RMAN automatically looks for a another usable copy of this backup piece or image copy, on the same device or another device, based on the information in the RMAN repository. If no usable copies are available, then RMAN searches for prior backups. RMAN searches all prior backups for the most recent available backup usable in the current operation until it has exhaused all possibilities.

Restore failover is also used when there are errors restoring archivelogs during RECOVER, BLOCKRECOVER, and FLASHBACK DATABASE commands.

When RMAN performs restore failover to another backup of the same file, you will see a message similar to this one in the output of RMAN:

failover to piece handle=/u01/backup/db_1 tag=BACKUP_031009

Also, details about block corruptions will be printed in the alert log and trace files.

When restore failover cannot locate another copy of the same backup and searches for a prior backup, the message generated is similar to this example:

ORA-19624: operation failed, retry possible
ORA-19505: failed to identify file "/u01/backup/db_1"
ORA-27037: unable to obtain file status
SVR4 Error: 2: No such file or directory
Additional information: 3
failover to previous backup

Restore Optimization

By default, RMAN does not perform a restore if the file to be restored is in the correct place and its header contains the expected information. RMAN only restores a file if the header check does not succeed, although you can use the FORCE option of the RESTORE command to override this behavior and restore the requested files unconditionally.

Restore optimization is particularly useful in cases where a restore only partially completes. For example, assume that a full database restore encounters a power failure after all except one of the datafiles has been restored. If you start the same restore again, then RMAN only restores the single datafile that was not restored during the previous attempt.

Datafile Media Recovery with RMAN

The concept of datafile media recovery is the application of online or archived redo logs or incremental backups to a restored datafile in order to update it to the current time or some other specified time. Use the RMAN RECOVER command to perform media recovery and apply logs or incremental backups automatically.

RMAN Media Recovery: Basic Steps

If possible, make the recovery catalog available to perform the media recovery. If it is not available, or if you do not maintain a recovery catalog, then RMAN uses metadata from the target database control file. If both the control file and recovery catalog are lost, then you can still recover the database--assuming that you have backups of the datafiles and at least one autobackup of the control file.

The generic steps for media recovery using RMAN are as follows:

1. Place the database in the appropriate state: mounted or open. For example, mount the database when performing whole database recovery, or open the database when performing online tablespace recovery.
2. To perform incomplete recovery, use the SET UNTIL command to specify the time, SCN, or log sequence number at which recovery terminates. Alternatively, specify the UNTIL clause on the RESTORE and RECOVER commands.
3. Restore the necessary files with the RESTORE command.
4. Recover the datafiles with the RECOVER command.
5. Place the database in its normal state. For example, open it or bring recovered tablespaces online.

Figure 3-1 illustrates an example of RMAN media recovery. The DBA runs the following commands:

RESTORE DATABASE;
RECOVER DATABASE;

RMAN then queries the repository, which in this example is a recovery catalog. The recovery catalog obtains its metadata from the target database control file. RMAN then decides which backup sets to restore, and which incremental backups and archived logs to use for recovery. A server session on the target database instance performs the actual work of restore and recovery.

Figure 3-1 Performing RMAN Media Recovery

Text description of the illustration bradv035.gif

Mechanics of Recovery: Incremental Backups and Redo Logs

If RMAN has a choice between applying an incremental backup or applying redo to the restored datafiles to meet a recovery objective, then it always chooses an incremental backup. If overlapping levels of incremental backup are available, then RMAN automatically chooses the one covering the longest period of time.

RMAN does not need to apply incremental backups to a restored level 0 incremental backup: it can also apply archived logs. RMAN restores the datafiles that it needs from available backups, applies incremental backups to the datafiles if they are available, and then applies archived logs.

How RMAN Searches for Archived Redo Logs During Recovery

If RMAN cannot find an incremental backup, then it looks in the repository for the names of archived redo logs to use for recovery. The database records an archived log in the control file whenever one of the following occurs:

• The archiver process archives a redo log
• RMAN restores an archived log
• The RMAN BACKUP AS COPY command copies a log
• The RMAN CATALOG command catalogs a user-managed backup of an archived log

If you use a recovery catalog, then RMAN propagates archived log data into the recovery catalog during resynchronization, classifying archived logs as image copies. You can view the log information through:

• The LIST ARCHIVELOG command
• The V$BACKUP_FILES control file view
• The V$ARCHIVED_LOG control file view

During recovery, RMAN looks for the needed logs using the filenames specified in the V$ARCHIVED_LOG view. If the logs were created in multiple destinations or were generated by the BACKUP AS COPY, CATALOG, or RESTORE commands, then multiple, identical copies of each log sequence number exist on disk. RMAN does not have a preference for one copy over another during recovery: all copies of a log sequence number listed as AVAILABLE are candidates. In a sense, RMAN is blind to the fact that the logs were generated in different destinations or in different ways.

If the RMAN repository indicates that a log has been deleted or uncataloged, then RMAN ceases to consider it as available for recovery. For example, assume that the database archives log 100 to directories /dest1 and /dest2. The RMAN repository indicates that /dest1/log100.arc and /dest2/log100.arc exist. If you delete /dest1/log100.arc with the DELETE command, then the repository indicates that only /dest2/log100.arc is available for recovery.

If the RMAN repository indicates that no copies of a needed log sequence number exist on disk, then RMAN looks in backups and restores archived redo logs as needed to perform the media recovery. By default, RMAN restores the archived redo logs to the first local archiving destination specified in the initialization parameter file. You can run the SET ARCHIVELOG DESTINATION command to specify a different restore location. If you specify the DELETE ARCHIVELOG option on RECOVER, then RMAN deletes the archived logs after restoring and applying them. If you also specify MAXSIZE integer on the RECOVER command, then RMAN restores archived logs until the disk space allowed by MAXSIZE is consumed, then applies redo from the logs and deletes the restored logs to free space, until there is room enough to restore another archived log. RMAN continues restoring, applying and deleting logs, within the MAXSIZE limit, until recovery is complete.

RMAN Behavior When the Repository Is Not Synchronized

If an archived log is deleted from disk and the repository does not reflect this fact, then RMAN does not perform automatic failover during recovery. For example, if the repository indicates that /dest1/log100.arc is on disk when in fact this log was deleted using an operating system command, and if RMAN attempts to apply this log file during recovery, then recovery terminates with an error. RMAN does not automatically attempt to apply other copies of log 100 that are listed as available in the repository.

This situation can sometimes occur when you delete an archived log with an operating system utility and then fail to run CROSSCHECK to synchronize the repository. If you run a CROSSCHECK so that the repository is synchronized, then recovery can proceed by applying available copies of the log or restoring a backup of the log if no disk copies are available.

Incomplete Recovery

RMAN can perform either complete or incomplete recovery. You can specify a time, SCN, or log sequence number as a limit for incomplete recovery with the SET UNTIL command or with an UNTIL clause specified directory on the RESTORE and RECOVER commands. The easiest method is run the SET UNTIL command before issuing the RESTORE and RECOVER commands. After performing incomplete recovery, you must open the database with the RESETLOGS option.

Tablespace Point-in-Time Recovery

Recovery Manager automated Tablespace Point-in-Time Recovery (TSPITR) enables you to recover one or more tablespaces to a point in time that is different from that of the rest of the database. RMAN TSPITR is most useful in these cases:

• To recover from an erroneous drop or truncate table operation
• To recover a table that has become logically corrupted
• To recover from an incorrect batch job or other DML statement that has affected only a subset of the database
• In cases where there are multiple logical schemas in separate tablespaces of one physical database, and where one schema must be recovered to a point different from that of the rest of the physical database
• For VLDBs (very large databases), even if a full database point-in-time recovery would suffice, you might choose to do tablespace point-in-time recovery rather than restore the whole database from a backup and perform a complete database roll forward

Similar to a table export, RMAN TSPITR enables you to recover a consistent data set; however, the data set is the entire tablespace rather than a single object.

Block Media Recovery with RMAN

Although datafile media recovery is the principal form of recovery, you can also use the RMAN BLOCKRECOVER command to perform block media recovery. Block media recovery recovers an individual corrupt datablock or set of datablocks within a datafile. In cases when a small number of blocks require media recovery, you can selectively restore and recover damaged blocks rather than whole datafiles.

Block media recovery provides several advantages over datafile media recovery. For example, block media recovery

• Lowers the Mean Time to Recovery (MTTR) because only blocks needing recovery are restored and only necessary corrupt blocks undergo recovery. Block media recovery minimizes redo application time and avoids I/O overhead during recovery.
• Allows affected datafiles to remain online during recovery of the blocks. Without block-level recovery, if even a single block is corrupt, then you must restore a backup of the entire datafile and apply all redo generated for that file after the backup was created.

Note these restrictions of block media recovery:

• You can only perform block media recovery with RMAN. No SQL*Plus recovery interface is available.
• You can only perform complete recovery of individual blocks. In other words, you cannot stop recovery before all redo has been applied to the block.
• You can only recover blocks marked media corrupt. The V$DATABASE_BLOCK_CORRUPTION view indicates which blocks in a file were marked corrupt since the most recent BACKUP or BACKUP ... VALIDATE command was run against the file.
• You must have a full RMAN backup. Incremental backups are not allowed. Note that Block media recovery is able to restore blocks from parent incarnation backups and recover the corrupted blocks through a RESETLOGS.
• Blocks that are marked media corrupt are not accessible to users until recovery is complete. Any attempt to use a block undergoing media recovery results in an error message indicating that the block is media corrupt.

  See Also: "Performing Block Media Recovery with RMAN" and Oracle Database Recovery Manager Reference for BLOCKRECOVER syntax

When Block Media Recovery Should Be Used

Block media recovery is not intended for cases where the extent of data loss or corruption is unknown and the entire datafile requires recovery. In such cases, datafile media recovery is the best solution. Block media recovery is not a replacement for traditional datafile media recovery, but a supplement to it.

In most cases, the database marks a block as media corrupt, invalidates the block in the instances (or all enabled instances in a Real Application Clusters configuration), and then writes it to disk when the corruption is first encountered. No subsequent read of the block will be successful until the block is recovered. You can only perform block recovery on blocks that are marked corrupt. This corrupt status effectively takes the block offline in all database instances and prevents user access during recovery.

Block media recovery is most useful for data losses that affect specific blocks. Block-level data loss usually results from intermittent, random I/O errors that do not cause widespread data loss, as well as memory corruptions that get written to disk. Typically, these types of block corruption are reported in the following locations:

• Error messages in standard output
• The alert log
• User trace files
• Results of the SQL commands ANALYZE TABLE and ANALYZE INDEX
• Results of the DBVERIFY utility
• Third-party media management output

For example, you may discover the following messages in a user trace file:

ORA-01578: ORACLE data block corrupted (file # 7, block # 3)
ORA-01110: data file 7: '/oracle/oradata/trgt/tools01.dbf'
ORA-01578: ORACLE data block corrupted (file # 2, block # 235)
ORA-01110: data file 2: '/oracle/oradata/trgt/undotbs01.dbf'

You can then specify the corrupt blocks in the BLOCKRECOVER command as follows:

BLOCKRECOVER
  DATAFILE 7 BLOCK 3 
  DATAFILE 2 BLOCK 235;

Block Media Recovery When Redo Is Missing

Like datafile media recovery, block media recovery cannot generally survive a missing or inaccessible archived log (although it will attempt restore failover when looking for usable copies of archived redo log files, as described in "Restore Failover"). Nevertheless, block media recovery can survive gaps in the redo stream if the missing or corrupt redo records do not affect the blocks being recovered. Whereas datafile recovery requires an unbroken series of redo changes from the beginning of recovery to the end, block media recovery only requires an unbroken set of redo changes for the blocks being recovered.

When RMAN first detects missing or corrupt redo records during block media recovery, it does not immediately signal an error because the block undergoing recovery may become a newed block later in the redo stream. When a block is newed all previous redo for that block becomes irrelevant because the redo applies to an old incarnation of the block. For example, the database can new a block when users delete all the rows recorded in the block or drop a table.

Assume that media recovery is performed on block 13 as depicted in Figure 3-2.

Figure 3-2 Performing RMAN Media Recovery

Text description of the illustration bradv007.gif

After block recovery begins, RMAN discovers that change 120 is missing. RMAN does not terminate recovery in the hope that block 13 will be newed later in the redo stream. Assume that in change 140 a user drops the table EMPLOYEE stored in block 13. At this point, the database formats block 13 as a new block. Because the redo for block 13 in change 120 related to the EMPLOYEE table, and the EMPLOYEE table was dropped in change 140, RMAN can skip this missing change and apply the redo between changes 140 and 160.

Database Duplication with RMAN

Use the RMAN DUPLICATE command to create a copy of the target database in another location. The command restores backups of the primary database files and creates a new database.

As part of the duplication, RMAN manages the following:

• Restores the target datafiles into the duplicate database and performs incomplete recovery using all available archived log and incremental backups
• Opens the duplicate database with the RESETLOGS option after incomplete recovery to create the online redo logs
• Generates a new, unique database identifier for the duplicate database

Note the following features of RMAN duplication. You can:

• Skip read-only tablespaces with the SKIP READONLY clause (read-only tablespaces are included by default). You can also exclude any tablespace with the SKIP TABLESPACE clause so long as it is not the SYSTEM or SYSAUX tablespace and does not contain rollback or undo data. If you omit tablespaces, then you can add them later.
• Create the duplicate database in a new host. If the same directory structure is available, then you can use the NOFILENAMECHECK option and reuse the target datafile filenames for the duplicate datafiles.
• Create the duplicate database by using the SET UNTIL command or UNTIL clause of the DUPLICATE command to recover it to a past time. By default, the DUPLICATE command creates the database using the most recent backups of the target database and then performs recovery to the most recent consistent point contained in the incremental and archived redo log backups.
• Use the duplicate database without a recovery catalog.
• Register the duplicate database in the same recovery catalog as the target database. This option is possible because the duplicate database receives a new database identifier during duplication. If you copy the target database with operating system utilities, then the database identifier of the copied database remains the same so you cannot register it in the same recovery catalog (unless you change its DBID with the DBNEWID utility, described in Oracle Database Utilities ).

Figure 3-3 illustrates a case of database duplication. In this example, RMAN creates two duplicate database by using one set of datafile backups: one database on the local host and one database on a remote host.

Figure 3-3 Creating a Duplicate Database from Backups

Text description of the illustration bradv027.gif

The method you use to duplicate your database depends on whether you are creating your duplicate database on the same or a different host and whether the duplicate directory structure is the same as your target database directory structure. For example, in some cases you can keep the same directory structure and filenames in your duplicate database, while other times you must rename the files.

Physical Standby Database Creation with RMAN

You can use the RMAN DUPLICATE command to create a physical standby database. (Note that RMAN cannot be used to create a logical standby database, because it is not a block-for-block duplicate of the primary database.) RMAN automates the following steps of the creation procedure:

1. Restores the standby control file.
2. Restores the primary datafile backups.
3. Optionally, RMAN recovers the standby database (after the control file has been mounted) up to the specified time or to the latest archived redo log generated.
4. RMAN leaves the database mounted so that the user can activate it, place it in manual or managed recovery mode, or open it in read-only mode.

RMAN cannot fully automate creation of the standby database because you must manually create an initialization parameter file for the standby database, start the standby instance without mounting the control file, and perform any Oracle Net setup required before performing the creation of the standby. Also, you must have RMAN backups of all datafiles available as well as a control file backup that is usable as a standby control file.

RMAN can back up the standby datafiles, control file and archived redo logs of a physical standby database. Backups of datafiles and archived redo logs taken from a physical standby database are fully interchangeable with primary backups. In other words, you can restore a backup of a physical standby datafile to the primary database, and you can restore a backup of a primary datafile to the physical standby database. The standby control file backups can be used to restore the standby control file without needing to re-instantiate the standby in cases where the standby control file is lost.