Novell ZENWORKS LINUX MANAGEMENT 7.3 IR2, Administration Manual

Page: 96 / 684

96 Novell ZENworks 7.3 Linux Management Administration Guide
no
vd
ocx
(e
n)
7 Ja
nua
ry 201
0
This results in all data files being restored and then recovered. RMAN applies archive logs as
necessary until the recovery is complete.
2
After the restore is complete, you need to ensure that the Data Store is synchronized with the
Object Store. For instructions, see
Section 9.4, “Synchronizing the Object Store and Data
Store,” on page 98 .
Incomplete Recovery
RMAN can perform recovery of the whole database to a specified non-current time, SCN, or log
sequence number. This type of recovery is called incomplete recovery because it does not
completely use all of the available redo logs. Incomplete recovery of the whole database is also
called database point-in-time recovery (DBPITR).
You should perform an incomplete recovery of the database in the following situations:
Š
Media failure destroys some or all of the online redo logs.
Š
A user error causes data loss, for example, a user inadvertently drops a table.
Š
You cannot perform a complete recovery because an archived redo log is missing.
To perform an incomplete recovery, restore all data files from backups created prior to the time
when a recovery is needed, and then open the database with the RESETLOGS option after recovery
completes. The RESETLOGS operation creates a new instance of the database—in other words, a
database with a new stream of log sequence numbers starting with log sequence 1.
The database must be closed to perform an incomplete recovery.
To perform an incomplete recovery:
1 Set the time format environment variable:
$ NLS_DATE_FORMAT="Mon DD YYYY HH24:MI:SS"
2
Use the following sequence of commands:
$ rman target /
RMAN> startup mount;
RMAN> run {
2> set until time "to_date('Jul 12 2009 10:24:00', 'MM DD YYYY
HH24:MI:SS')";
3> restore database ;
4> recover database ;
5> }
RMAN uses the last backup created before the time mentioned in the set until command to
restore the files to their default locations. Then, it uses archived redo logs (if needed) to recover
the database.
Two other parameters that can be used with the
set until
command are SCN and log
sequence numbers. You obtain SCNs from the alert logs. Find the SCN of an event and recover
to a prior SCN. For example:
SET UNTIL SCN 1000
3 If recovery was successful, open the database and reset the online logs: