Datawarehouse backup

All:

We are planning to cold backup for 1 TeraBytes of datawarehouse database on weekly basis (every Saturday). Version :9i R2,
OS: Sun Solaris
No Archivelog Mode

These are the steps we decide to perform.

A)Backup

1. Stop ETL jobs
2. Change all the tablespaces to READ-ONLY mode except SYSTEM
3. Alter system switch logfile( 3 times)
4. Shutdown database
5. Copy system tablespace datafiles, the control files and the online redo logs.
6. Startup database
7. Compress & Copy all the READ-ONLY tablespaces datafiles to tape
8. Change all the READ-ONLY tablespaces to READ-WRITE
9. Start ETL job
10. Recovery:
11. Backup previous cold backup
12. Run ETL jobs up to date

The only db outage would be between step # 3 and 5. These steps will be controlled from control-M product.

I would like to know, is there any problem in this approach or is there any better way to perform the same?

Thanks in advance
--Sami
--

http://www.freelists.org/webpage/oracle-l Received on Wed Mar 09 2005 - 22:44:44 CST