Restore and Recover an Entire Database in NOARCHIVELOG Mode to a Point-In-Time

The following procedure describes the steps involved in restoring and recovering an entire database when all data files and control files are lost. This procedure restores and recovers an entire database as follows:

• From an offline backup

• With the control files (since they no longer exist)

• To the latest backup and to the original host

• Using the Recovery Catalog

  When you are restoring data from a database in NOARCHIVELOG mode, you must always restore and recover the entire database.

Before You Begin:

• Review the general and agent-specific restore requirements accessed from Restore Backup Data prior to performing any restore.

Required Capability: See Capabilities and Permitted Actions

To restore and recover an entire database in NOARCHIVELOG mode to a point-in-time:

1. From the CommCell Browser, right-click the instance that contains the database you want to restore and recover, click All Tasks and then click Restore.
2. From the Oracle Restore Options (General) dialog box, select the following options:
   • Select the name of the client computer with the database you want to restore in the Destination Client box (if it is not already selected).
   • Select the Restore Control File option.
   • Select the Restore Data and Recover options.

     It is critical that the Recover option is selected, as Recover applies any incremental backups that are needed to bring the database back to a consistent state.

   • Select the entire database in The Latest Database View pane (if it is not already selected).
   • Verify that the Status of the database is displayed as STARTED; if necessary click on the Refresh button to refresh the status.
3. Click the Advanced button and verify the following options are selected in the identified tab for the Oracle Advanced Restore Options dialog box:
   • Restore Control File - To Point in Time in Ctrl & SP Files

     The Control file must be restored to a point in time that is later than or equal to time set for the Restore Data option in the Restore tab.

   • Restore Data in Restore

     The Restore Data option must be earlier than or equal to the point in time that you specified for the Restore Control File option in the Ctrl & SP Files tab.

   • Recover - Point In Time in Recover

     The point in time for the recover option must be later than or equal to the point in time specified for the Restore Control File in the Ctrl & SP Files tab.

     If you recover the database to a previous point in time and restore the control file, then you have to perform a full database backup. This is because a previous point in time recovery and restoring the control file will destroy all previous backups.

4. Click OK in the Oracle Advanced Restore Options window.
5. When restoring encrypted data, refer to Data Encryption.
6. Optionally for Oracle RAC, from the Oracle Restore Options (Stream) dialog box, set the instance restore order per stream allotment.
7. Begin or schedule the restore.

   Before the restore job is run, the system will check whether the database is in the correct mode for this type of restore. If it is not, a message dialog will inform you of this and offer to automatically switch the database into the valid mode. When this happens, either click Yes on the message dialog to have the system automatically switch the database to the valid mode and continue the restore, or click No to manually switch the database to the valid mode before proceeding.

8. View the progress of the restore in the Job Controller window.
9. The job will report the restore as failed. This is because no archive redo log files exist. RMAN will report the following message, in reference to this error:

   RMAN-00569: =========error message stack follows============
   RMAN-03002: failure during compilation of command
   RMAN-03013: command type: recover
   RMAN-06038: recovery catalog package detected an error
   RMAN-03013: command type: recover(4)
   RMAN-20242: specification does not match any archivelog in the recovery catalog

   Manually recover and open the database using the following steps.

10. Make sure that ORACLE_SID is set to the database you are restoring and that ORACLE_HOME is set appropriately. Note the following example:
    #export ORACLE_SID= <target database SID>
    #export ORACLE_HOME= <Oracle home directory>
11. From the Server Manager window, execute the command displayed in the following example:
    SVRMGR>recover database using backup control file until cancel;

    Server Manager displays several messages, including the following:
    Specify log: {<RET>=suggested | filename | AUTO | CANCEL}

12. Type cancel

    The following message is displayed:
    Media recovery cancelled.

13. Execute the commands displayed in the following example:
    SVRMGR>alter database open resetlogs;
    SVRMGR>exit
14. Reset the Recovery Catalog by executing the commands displayed in the following example:
    rman rcvcat <catalog connect string>
    Recovery Manager: Release 8.1.5.00 - Production
    RMAN-06008: connected to recovery catalog database
    RMAN>connect target;
    RMAN-06005: connected to target database: <database_name> DBID=<##>)
    RMAN> reset database;
    RMAN-03022: compiling command: reset
    RMAN-03023: executing command: reset
    RMAN-08006: database registered in recovery catalog
    RMAN-03023: executing command: full resync
    RMAN-08002: starting full resync of recovery catalog
    RMAN-08004: full resync complete
    RMAN>exit

    Resetting the Recovery Catalog creates a new incarnation of the database. Before you can restore and recover the database to the current time, you must perform another full backup and wait for it to complete successfully.

15. Perform an immediate offline full backup.