RMAN backup types

1-A whole database backup includes all datafiles and at least one control file.
Partial database backups may include zero or more tablespaces, zero or more datafiles, and
may or may not include a control file.
Full backups are non-incremental RMAN backups. You can make a full backup of only one
datafile. “Full” does not refer to how much of the database is backed up, but to the fact that the
backup is not incremental.

2-Incremental backups make a copy of all data blocks that have changed since some previous
backup. Oracle Database 10g supports two levels of incremental backup (0 and 1). A level 0 or
baseline backup is equivalent to a full backup and contains all data blocks. A level 1
cumulative incremental backs up all database blocks changed since the level 0 backup. A level
1 differential incremental backs up all database blocks changed since a level 0 or level 1
backup. To restore using incremental backups, the baseline backup must first be restored, and
then the incremental backup. You can make incremental backups with Recovery Manager
(RMAN).
Offline backups (also known as cold or consistent backups) are taken while the database is not
open. They are consistent because at the time of the backup, the system change number (SCN)
in the datafile headers matches the SCN in the control files.


Online backups (also known as hot or inconsistent backups) are taken while the database is
open. The backups are inconsistent because, with the database open, there is no guarantee that
the datafiles are synchronized with the control files.

Why Do we need to backup our data?

We need backup because of the following:
• Media failure: The failure of a read or write of a database file due to a physical problem with the file.
• User errors: Data in your database is changed or deleted incorrectly.

Responding to User Error
Possible responses to user error include:
• Re-entering the lost data manually, if a record of the changes exists.
• Importing the dropped object, if a suitable export file exists or the object is still available on a standby
database.
• Performing tablespace point-in-time recovery (TSPITR) of one or more tablespaces.
• Returning the database to a past state using database point-in-time recovery.
• Using one of the Oracle Flashback features to recover from logical corruption by returning affected objects to a past state.