Using LogMiner
Oracle uses online redo log files to track every change that is made to user data and the data
dictionary. The information stored in the redo log files is used to re-create the database, in part
or in full, during recovery. To enable recovery of the database to a point in time after the database
backup was made, you can maintain archived copies of the redo log files. The LogMiner utility
provides a vital view into the modifications that have occurred within your database.
When you use LogMiner, you see both the changes that have been made (the SQL_redo) and
the SQL you can use to reverse those changes (the SQL_undo). Thus, you can review the history
of the database, without actually applying any redo logs, and obtain the code to reverse any
problematic transactions. Using LogMiner, you can pinpoint the transaction under which corruption
first occurred so that you can determine the correct point in time or System Change Number (SCN)
to use as the endpoint for a database recovery.
If there were a small number of transactions that required rolling back, prior to LogMiner, you
would have to restore the table to an earlier state and apply archived log files to bring the table
forward to just before the corruption. When restoring the table and applying the archived log files,
you would risk losing later transactions that you would like to retain. You can now use LogMiner
to roll back only the transactions that are problematic without losing later, valid transactions.
Pages:
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757