Recovery practice

Hello group!

Trying to understand how to recover after some types of disasters.

When I delete all control files, all log files and all database files at the same time- I'm fried...

I can't recreate the control file since the data files are missing, and I can't restore data files since the db doesn't mount without a control file...

The "backup" of the controlfile was taken by using SQL> ALTER DATABASE BACKUP CONTROLFILE TO TRACE; If I instead had used
SQL> ALTER DATABASE BACKUP CONTROLFILE TO '/oracle/backup/control.bkp';

could I have done
$ cp /oracle/backup/control.bkp
/u01/app/oracle/oradata/test/control01.ctl and been saved?

I need better understand the nature of control files, right?

I'm reading
Backup and Recovery Basics (B10735-01)
Backup and Recovery Advanced User's Guide (B10734-01)

The test system is White Box Linux and Oracle 10g in archivelog (using the flashback area). This rman script is run weekly:

---

connect target;
run {
  RECOVER COPY OF DATABASE WITH TAG "S";   BACKUP INCREMENTAL LEVEL 1

        FOR RECOVER OF COPY WITH TAG "S"
        DATABASE;

And finally, no I'm not a DBA- I'm just a poor sysadm that was told to keep an eye on Oracle meanwhile the company finds a new one [DBA]...

Thanks,
ph Received on Fri Apr 15 2005 - 14:32:30 CDT