Databases Reference
In-Depth Information
3.
Create an incremental backup of the primary database from the current SCN, as shown here:
RMAN> backup incremental from scn 123456 database
format '/tmp/ForCatchup_%U' tag 'ForCatchp';
The RMAN
backup
command shown in step 3 will create data file backups and a control
file backup. Move all the backup pieces that you've created on the primary database server
to the standby database server:
4.
$ scp /tmp/Forcatchup_* standby:/tmp
5.
On the standby database, execute the following command to catalog the backup pieces
you moved from the primary server to the standby server:
RMAN> catalog start with '/tmp/ForCatchp';
6.
Check that the file names at the standby database site are registered, by executing the
following command:
RMAN> report schema;
7.
Apply the incremental backups to the standby database by executing the following set of
commands:
RMAN> connect target /
RMAN> startup force nomount;
RMAN> restore standby controlfile from tag 'FORSTANDBY';
RMAN> alter database mount;
RMAN> recover database noredo;
8.
Start the apply of redo on the physical standby:
SQL> alter database recover managed standby database disconnect from session;
How It Works
You can sometimes use RMAN incremental backups to synchronize a standby with the primary database. First
create a backup on the primary database using the RMAN
backup incremental from SCN
command. The backup
starts at the current SCN of the standby, so you can then roll the standby database forward in time by applying this
incremental backup from the primary database.
You can use this strategy of applying RMAN incemental backup in the following situations:
•
The physical standby lags far behind the primary database.
•
The physical database has a large number of nologging changes.
•
Several data files in the physical standby have nologging changes.
Although a recovery catalog isn't absolutely essential to follow the incremental backup strategy described here,
Oracle recommends you use the recovery catalog so you don't have to correct the file names in the restored control file.
In the example shown in this recipe, we used the current SCN of the standby database as the starting point of the
primary database backup. However, if you're dealing with a standby with a large number of nologging changes, you
must use the
first_nologged_scn
from the V$DATAFILE view, as shown here:
