Enabling media recovery in 10g

Hi All,

I'm trying to understand the functionality of the "alter database begin backup" command in 10g. When i issue this command with a test database, i get the following error: ORA-01123: cannot start online backup; media recovery not enabled.

How do i enable media recovery?

Thanks in advance for your help!

Hi

I would suggest you read the oracle concepts guide.

regards
Hrishy

To enable archivelog mode so that Oracle saves the old redo logs to disk after a log switch before overwriting them when it cycles back around, perform the following commands. Oracle requires at least two redo logs (log groups) in order to accommodate possible archiving, since one could be being spooled during archiving to the archivelogs directory while changes currently being made to the database are being written to the other by the log writer.

$ vi $ORACLE_HOME/dbs/initPROD.ora
Edit init.ora file to contain the archive parameters (directory name and file format and the start flag):
log_archive_dest = /u01/oradata/PROD/archivelogs/
log_archive_format = arch_PROD_%S.arc
log_archive_start = true
("archive log start" can also be used when connected as sysdba to start the arch process, if you don't want it automatically done through init.ora; but, neither of these turn on archivelog mode - they just start the arch process itself.)
$ mkdir /u01/oradata/PROD/archivelogs
$ sqlplus "/ as sysdba"
SQL> shutdown immediate
SQL> startup mount
SQL> alter database archivelog;
Turns on archivelog mode for the database.
SQL> alter database open;
SQL> archive log list
Shows that you are now in archivelog mode, that archiving is enabled, and shows you the current log sequence numbers. You can "alter system switch logfile;" a few times to see archivelog files going into the log_archive_dest directory.
SQL> select * from v$database;
Also shows that you are in archivelog mode.

Quote:

---

Originally Posted by Kailash Chandra

To enable archivelog mode so that Oracle saves the old redo logs to disk after a log switch before overwriting them when it cycles back around, perform the following commands. Oracle requires at least two redo logs (log groups) in order to accommodate possible archiving, since one could be being spooled during archiving to the archivelogs directory while changes currently being made to the database are being written to the other by the log writer.

$ vi $ORACLE_HOME/dbs/initPROD.ora
Edit init.ora file to contain the archive parameters (directory name and file format and the start flag):
log_archive_dest = /u01/oradata/PROD/archivelogs/
log_archive_format = arch_PROD_%S.arc
log_archive_start = true
("archive log start" can also be used when connected as sysdba to start the arch process, if you don't want it automatically done through init.ora; but, neither of these turn on archivelog mode - they just start the arch process itself.)
$ mkdir /u01/oradata/PROD/archivelogs
$ sqlplus "/ as sysdba"
SQL> shutdown immediate
SQL> startup mount
SQL> alter database archivelog;
Turns on archivelog mode for the database.
SQL> alter database open;
SQL> archive log list
Shows that you are now in archivelog mode, that archiving is enabled, and shows you the current log sequence numbers. You can "alter system switch logfile;" a few times to see archivelog files going into the log_archive_dest directory.
SQL> select * from v$database;
Also shows that you are in archivelog mode.

---

you need to get with the times ad work out what is and what isnt required for a 10g database

You also he is on some unix systemm not much use if you are on windows is it?