Database Reference
In-Depth Information
Step 1. Shut Down the Instance
Shut down the instance, if it's running. In this scenario it doesn't matter how you shut down the database, because
you're restoring back to a point in time (with no recovery of transactions). Any files in the live database directory
locations are overwritten when the backup files are copied back. If your instance is running, you can abruptly abort it.
As a
SYSDBA
-privileged user, do the following:
$ sqlplus / as sysdba
SQL> shutdown abort;
Step 2. Copy the Files Back from the Backup
This step does the reverse of the backup: you're copying files from the backup location to the live database file
locations. In this example all the backup files are located in the
/u01/cbackup/O12C
directory, and all files are being
copied to the
/u01/dbfile/O12C
directory:
$ cp /u01/cbackup/O12C/*.* /u01/dbfile/O12C
Step 3. Start Up the Database
Connect to your database as
SYS
(or a user that has
SYSDBA
privileges), and start up your database:
$ sqlplus / as sysdba
SQL> startup;
After you finish these steps, you should have an exact copy of your database as it was when you made the cold
backup. It's as if you set your database back to the point in time when you made the backup.
Restoring a Cold Backup in Noarchivelog Mode Without Online Redo Logs
As mentioned earlier, you don't ever need the online redo logs when restoring from a cold backup. If you made a cold
backup of your database in noarchivelog mode and didn't include the online redo logs as part of the backup, the steps
to restore are nearly identical to the steps in the previous section. The main difference is that the last step requires you
to open your database, using the
OPEN RESETLOGS
clause. Here are the steps:
1.
Shut down the instance.
2.
Copy the control files and data files back from the backup.
3.
Start up the database in mount mode.
Open the database with the
OPEN RESETLOGS
clause.
4.
Step 1. Shut Down the Instance
Shut down the instance, if it's running. In this scenario it doesn't matter how you shut down the database, because
you're restoring back to a point in time. Any files in the live database directory locations are overwritten when
the backups are copied. If your instance is running, you can abruptly abort it. As a
SYSDBA
-privileged user, do the
following:
$ sqlplus / as sysdba
SQL> shutdown abort;
