Databases Reference
In-Depth Information
How It Works
The ability to encrypt backup sets during active database duplication provides additional security when transmitting
the contents of the data files over a network. You can specify encryption during active database duplication, but only if
you're using push-based duplication, which uses RMAN backup sets and not image copies. Since backup sets can be
used in database duplication only starting with Oracle Database 12.1 release, you can specify encryption only if your
database is atleast 12.1 or newer releases.
Using compressed backup sets will provide better performance during the duplication of a database, because
you'll be dealing with smaller backup sets. RMAN supports the compression of the backupsets used for active
database duplication starting with Oracle Database 12c Release 1.
If you're dealing with very large source database data files, using multisection backups during active database
duplication provides faster performance during the backup. RMAN will back up the large data files in parallel and
create multiple backup pieces with separate RMAN channels dedicated to creating each backup piece. As with the
encryption and compression features, the ability to create multisection backups is also a new feature introduced in
the 12.1 release.
Duplicating a Database with Several Directories
Problem
You want to duplicate a source database with data spread out over several directories.
Solution
If the source database files are spread over multiple directories, you must use the
set newname
parameter instead of
the
db_file_name_convert
parameter to rename the files in the duplicate database.
The following example shows how to create a duplicate database when the target database uses several
directories:
RMAN> run
{
set newname for datafile 1 to '/u01/app/oracle/testdata/system01.dbf';
set newname for datafile 2 to '/u01/app/oracle/testdata/sysaux01.dbf';
set newname for datafile 3 to '/u01/app/oracle/testdata/data01.dbf';
set newname for datafile 4 to '/u01/app/oracle/testdata/index01.dbf';
set newname for datafile 5 to '/u01/app/oracle/testdata/undotbs01.dbf';
duplicate target database to newdb
logfile
group 1 ('/u01/app/oracle/testdata/logs/redo01a.log',
('/u01/app/oracle/testdata/logs/redo01b.log') size 10m reuse,
group 2 ('/u01/app/oracle/testdata/logs/redo02a.log',
('/u01/app/oracle/testdata/logs/redo02b.log') size 10m reuse;
}
The
run
block shown here duplicates the target database and supplies file names for both data files and online
redo log files for the duplicate database.
