Oracle X6-2-HA Deployment And User's Manual page 147

• SCN: SCN-based recovery that performs a recovery is an incomplete recovery to
a specified SCN within the current incarnation of database.
• BackupReport: Performs a recovery based on the Backup Report timestamp. The
BackupReport option is similar to PITR, except that the recovery uses the Backup
Report timestamp instead of a specified timestamp.
Point-in-Time Database Recovery
Point-in-Time Recovery (PITR) recovers the database to a specific point-in-time. You
can use a timestamp or a backup report for this type of recovery.
A point-in-time recovery is often used for the following situations:
• You want to recover the database to an SCN before a user or administrative error.
• Complete database recovery failed because all necessary archived redo logs were
not available.
• A database upgrade fails.
• You are creating a test database or a reporting database from production
database backups.
System Change Number Recovery
A system change number (SCN) is a stamp that defines a committed version of a
database at a point in time. Oracle assigns every committed transaction a unique
SCN. You can recover a database to the latest time or to an SCN.
The database is restored from backups created before the target time, and then
applies incremental backups and logs to re-create all changes between the time of the
data file backups and the end point of recovery. When the end point is specified as an
SCN, the database applies the redo logs and stops after each redo thread or the
specified SCN, whichever occurs first. When the end point is specified as a time, the
database internally determines a suitable SCN for the specified time and then recovers
to this SCN.
SCN is a primary mechanism to maintain data consistency in Oracle database. With
each execution, there is an increase in the count of the SCN. The following are some
examples of areas that use SCN:
Redo records. Every redo record has an SCN version of the redo record in the
1.
redo header (redo records can have non-unique SCN). Given redo records from
two threads (as in the case of RAC), recovery will order them in SCN order,
essentially maintaining a strict sequential order.
Data block. Every data block has block SCN (block version). In addition to that, a
2.
change vector in a redo record also has expected block SCN. This means that a
change vector can be applied to one and only version of the block. Code checks if
the target SCN in a change vector is matching with the block SCN before applying
the redo record. If there is a mismatch, corruption errors are thrown.
Read consistency. Every query has query environment which includes an SCN at
3.
the start of the query. A session can see the transactional changes only if that
transaction commit SCN is lower then the query environment SCN.
Commit. Every commit generates an SCN (commit SCN) that marks a transaction
4.
boundary. Group commits are possible too.
About Database Backup and Recovery Options
Chapter 11
11-7