In Recovery Manager (RMAN), you are taking image copies of the data files of your
production database and rolling them forward at regular intervals. You attempt to restart your
database. After a regular maintenance task, you realize that one of the data files that
belongs to the USERS tablespace is damaged and you need to recover the data file by
using the image copy. Because a media failure caused the data file to be damaged, you
want to place the data file in a different location while restoring the file.
Which option must you consider for this task?
A.
using only the RMAN SWITCH command to set the new location for the data file
B.
placing the database in the MOUNT state for the restore and recovery operations.
C.
using an RMAN RUN block with the SET NEWNAME and then the SWITCH command.
D.
configuring two channels: one for the restore operation and the other for the recovery
operation
I guess the C answer makes sense when you consider that A says:
using only the RMAN SWITCH command to <>
But when we’re using image copies, isn’t better to use switch datafile to copy?
That’s guessing the location where the image copies reside it’s not compromised
Answer is C
Here is an example of recovering a non critical datafile …
RMAN> RUN {
sql ‘ALTER DATABASE DATAFILE 4 OFFLINE’; — to run a SQL*Plus command in RMAN
SET NEWNAME FOR DATAFILE 4 TO ‘/u02/df_copies/users.dbf’;
SWITCH DATAFILE 4; — updates the control file with the new details for DATAFILE 4
RECOVER DATAFILE 4; — or use ALL
sql ‘ALTER DATABASE DATAFILE 4 ONLINE’;
}
Note: this can be done while the rest of the database is ONLINE. (Good for High Availability)