Databases Reference
In-Depth Information
If you'd rather create a standby database instead of a duplicate database, all you have to do is replace the
to
auxdb
part of the
duplicate database
command with the
for standby
clause, as shown here:
RMAN> duplicate target database
for standby
from active database;
It's easy to duplicate a non-ASM file-based database to an ASM file system-based database. Here we'll show you
a simple example to demonstrate how to do this. First create an ASM disk group, named +DISK1. Following is the
database duplication command to create an ASM file system-based duplicate database. This
duplicate database
command will create a database whose data files, control files, and online redo logs are all in the ASM disk group
+DISK1
.
RMAN> duplicate target database
to newdb
from active database
spfile
parameter_value_convert
'/u01/app/oracle/oradata/sourcedb/','+DISK1'
set db_create_file_dest = +DISK1;
You can speed up active database duplication by increasing the parallelism setting of disk channels on the source
database. Doing so will lead to parallel copying of the source database files over the network.
■
Make sure you don't use the
nofilenamecheck
clause with the
duplicate
command when duplicating a
database on the same host as the source database, as is the case in this recipe. the
nofilenamecheck
clause keeps
rMan from checking whether the source and target file names are identical. use it only when duplicating a database to
a different host.
Tip
If you get an error such as PLS-561, it means the duplicate database doesn't have the same value for the
nls_lang
parameter as the source database. In this case, you must set the correct
nls_lang
environment variable to get past the
error. Use the following query to find out the value of the
nls_lang
parameter in the source database:
SQL> select parameter, value from nls_database_parameters
where parameter in
('NLS_LANGUAGE','NLS_TERRITORY','NLS_CHARACTERSET'
,'NLS_NCHAR_CHARACTERSET');
To change the value of the
nls_lang
variable, use the
export
or
setenv
command (in Linux/Unix) to set the value
for the variable you get from the previous query.To avoid problems because of inadequate memory, you may use
a higher memory for the duplicate database when you're duplicating it.
