Skip to main content
NetApp Backup and Recovery

Restore Oracle Database workloads with NetApp Backup and Recovery

Contributors netapp-mwallis
PDFs

Restore Oracle Database workloads from snapshots, replicated backups on secondary storage, or backups stored in object storage by using NetApp Backup and Recovery.

Restore from these locations

You can restore databases from different starting locations:

  • Restore from a primary location (local snapshot)

  • Restore from a replicated resource on secondary storage

  • Restore from an object storage backup

Restore to these points

You can restore a database to its original location, or restore it as a new database on a different registered host (alternate location).

  • Restore to the original location

  • Restore to an alternate location (a different registered host)

How Oracle Database restore works

When you restore Oracle Database workloads, the following occurs:

  • When you restore a database from a local snapshot, NetApp Backup and Recovery creates a new resource using the data from the backup.

  • When you restore from replicated storage, you can restore it to the original location.

  • When you restore to an alternate location, NetApp Backup and Recovery restores the selected backup of the source database as a new database on a different registered host. You specify the target host and a name for the new database. If you don't provide a restore specification file, Backup and Recovery automatically calculates the locations for the data files, control files, and redo log files on the target host.

  • When you restore a backup from object storage, you can restore the data to the source storage or to an on-premises ONTAP system and recover the database from there.

View information about available Oracle Database restore points

View the available restore points, including the SCN number and associated database incarnation.

Note Oracle Database creates a new database incarnation each time you open a database with the RESETLOGS option. NetApp Backup and Recovery lists the relevant database incarnations when you review restore points and when you restore to a specific SCN.
Steps

  1. From the Backup and Recovery menu, select Inventory.

  2. From the NetApp Backup and Recovery menu, select Restore.

  3. Select View for the Oracle Database workload.

  4. Select the Databases tab.

  5. Select the Actions Actions icon icon for a database whose restore points you want to view, and select View protection details.

  6. Review the available restore points and the SCN number and database incarnation associated with each one.

Restore an Oracle database

Depending on your needs, restore an Oracle database to a specific point in time, to a specific system change number (SCN), or to the last good state. You can also restore the database from snapshots and skip the automated recovery process, for example, if you want to perform recovery manually. You can search for the database using its name or with specific filters.

Required NetApp Console role
Backup and Recovery super admin or Backup and Recovery restore admin role. Learn about NetApp Console access roles for all services.

Before you begin

When you restore to an alternate location, keep in mind the following supported and unsupported restore scenarios:

Supported alternate location restore scenarios

  • Restoring a single, whole Oracle database to a single target host

  • Recovery to a specific point in time, to a specific SCN, or to the latest backup

  • Using external archive log directories when the archive logs required for the selected recovery point are not included in the backup

Unsupported alternate location restore scenarios

  • Tablespace-only restore, including restore of tablespaces within a PDB

  • PDB-only restore to a different container database (CDB)

  • Restore from secondary storage

  • Restore from object storage

Note

If restoring PDBs or tablespaces to a specific point in time or SCN:

  • Select only one PDB. Point-in-time recovery is not compatible with restoring multiple PDBs.

  • PDB$SEED cannot be selected for point-in-time recovery.

  • Control file restore cannot be combined with point-in-time recovery of PDBs or PDB tablespaces.

  • The SYSTEM, SYSAUX, and UNDO tablespaces cannot be selected.

Important notes about restoring Oracle Data Guard databases

  • Before you restore, make sure the database role is current. If a switchover or failover has happened, select the Hosts tab and select Refresh for the host first. The available restore options depend on the role the database has now.

  • If the database is currently a standby, you can't restore it to the original location, regardless of the role it had when the backup was taken. Restore it to an alternate location.

  • If the database is currently a standby, you can't restore a snapshot that exists only on secondary storage or in object storage.

  • You can restore a backup taken while the database was a standby to the original location after the database becomes the primary.

Caution Restoring a primary database doesn't change the standby. When you open the primary with RESETLOGS after a point-in-time or SCN restore, it starts a new database incarnation. If the standby has already applied redo past that point, redo apply on the standby stops with an error and the standby stays at its earlier state. To resume protection, re-establish the standby after the restore by using Oracle tools.
Steps

  1. From the NetApp Console menu, select Protection > Backup and Recovery.

  2. From the NetApp Backup and Recovery menu, select Restore.

  3. From the drop-down list to the right of the name search field, select Oracle.

  4. Enter the name of the database you want to restore or filter for the database host where the database that you want to restore is located.

    A list of snapshots appears that match your search criteria.

  5. Select the Restore button for the database that you want to restore.

  6. Choose a restore option:

    Restore to specific point in time

    1. Select Restore to specific point in time.

    2. Select Next.

    3. Choose a date from the dropdown, and select Search.

      The closest snapshot to the specified date and time is displayed.

    4. If the snapshot is stored in more than one location, select a source location to use and add or review location details.

    5. Select Next.

    Restore to a specific system change number (SCN)

    1. Optionally, refer to View information about available Oracle Database restore points to see the available SCNs and associated database incarnations.

    2. Select Restore to a specific system change number (SCN).

    3. Select Next.

    4. Enter the SCN to use as a restore point, and select Search.

      The closest matching snapshot for the specified SCN is displayed. If multiple database incarnations match the SCN, all matching incarnations are shown.

    5. If multiple database incarnations match the search, select the snapshot with the correct incarnation.

    6. If the snapshot is stored in more than one location, select a source location to use and add or review location details.

    7. Select Next.

    Restore to the latest backup (last good state)

    1. Select Restore to the latest backup.

    2. Select Next.

      The latest snapshot is displayed. The restore operation recovers the last good state using all available log backups.

    3. If the snapshot is stored in more than one location, select a source location to use and add or review location details.

    4. Select Next.

    Restore from snapshots with no recovery

    1. Select Restore from snapshots with no recovery.

      Selecting this option means that the database will be restored from snapshots, but recovery is skipped (log backups are not applied).

    2. Select Next.

      Available snapshots are displayed.

    3. Select a snapshot from the list and select Next.

    4. If the snapshot is stored in more than one location, select a source location to use and add or review location details.

    5. Select Next.

  7. Choose the restore destination and settings:

    Restore to original location

    1. In the Destination settings section, do the following:

      1. For RAC databases, select the RAC node on which to run the restore. This option appears only for RAC databases.

      2. Select what you want to restore:

        • Select All datafiles to restore the entire database or container database (CDB).

        • Select Pluggable databases (PDBs) to restore one or more PDBs from a CDB without affecting the other PDBs in the container. Then select the PDBs from the list.

        • Select Pluggable database (PDB) tablespaces (shown as Tablespaces for a non-CDB database) to restore specific PDB tablespaces. Then, select the PDB and the tablespaces.

      3. Optionally, enable the Control files option to also restore the database control files. This option cannot be selected for point-in-time or SCN recovery of PDBs or PDB tablespaces.

      4. Optionally, enable the Change database state if needed for restore and recovery option to allow Backup and Recovery to manage the database state depending on the type of recovery.

      5. Optionally, enable the Force in place restore option to perform an in-place restore when new datafiles were added after the backup, or when LUNs were added, deleted, or re-created in an LVM disk group. This overrides the foreign-files check; database recovery proceeds using the restore option you selected.

      6. Specify external archive log file locations in the Specify external archive log files locations field if restoring a database to a specific point in time or SCN, and the archive logs are not in the configured location. Enter a file system path that starts with / or an ASM disk group that starts with +. Use the "plus" and "minus" buttons next to the text entry field to add or remove locations. This field does not appear when you restore from snapshots with no recovery.

      7. Optionally, enable the Open the database or container database in READ-WRITE mode after recovery option and select a sub-option that matches the restore scope (these options don't appear when restoring from snapshots with no recovery):

        • All datafiles: Open the database or container database in READ-WRITE mode after recovery.

        • PDBs or PDB tablespaces: Open the pluggable databases in READ-WRITE mode after recovery.

        • Non-CDB tablespaces: Bring the tablespaces online after recovery.

    2. In the Prescripts and postscripts section, do the following:

      1. Prescript: Optionally, enable this option and enter the full path for a script that should be run before the restore operation and any arguments that the script takes.

      2. Postscript: Optionally, enable this option and enter the full path for a script that should be run after the restore operation and any arguments that the script takes.

      3. Choose a timeout value for the script. If the script fails to execute within this time period, the restore will proceed anyway.

    Restore to alternate location

    Restore a database as a new database on a different registered host. Review the supported and unsupported scenarios before you begin.

    1. In the Destination settings section, enter the following information:

      1. Target host: Select the registered host where you want to create the new database.

      2. Target database: Enter a name for the target database. By default, the name starts with the name of the source database, and you can edit it.

      3. Customize target database parameters: Enable this option to upload a custom restore specification file to customize the target database parameters.

      4. Restore specification file: If you enabled the Customize target database parameters option, provide a restore specification file to control where the restored files are placed on the target host. You can download a default restore specification file and customize it to change the target database details such as authentication and log locations.

        If you don't provide a restore specification file, Backup and Recovery automatically calculates the locations for the data files, control files, and redo log files.

      5. Specify external archive log file locations in the Specify external archive log files locations field if restoring a database to a specific point in time or SCN, and the archive logs are not in the configured location. Enter a file system path that starts with / or an ASM disk group that starts with +. Use the "plus" and "minus" buttons next to the text entry field to add or remove locations. This field does not appear when you restore from snapshots with no recovery.

      6. Optionally, enable the Open the database or container database in READ-WRITE mode after recovery option and select a sub-option that matches the restore scope (these options don't appear when restoring from snapshots with no recovery):

  8. Select Restore.

    After the restore operation completes, check the Inventory page to confirm that the new database is listed. If the database is not listed, refresh the target host.