I've implemented a backup (and recovery) plan for an Oracle 11gR2 database as follows:
- Installed the same version of 11gR2 on production and disaster recovery (DR) servers
- Created identical Oracle instances (filesystem structures, SID, etc.) on both servers
- Every 6 hours I'm performing a hot backup on the production system and copying the files to the DR system:
alter database begin backup;
<copy all data files to DR server>
alter database end backup;
alter system archive log current;
alter database backup controlfile to 'c:\temp\control01.ctl';
alter database backup controlfile to trace as 'c:\temp\create_control.sql';
<copy both control files to DR server>
<copy all archived logs to DR server>
- When I want to recover on the DR system I can simply make sure all the files are in the correct places, "startup mount", "recover automatic database until cancel using backup controlfile;", and "alter database open resetlogs". At this point I only need to change a few dba passwords and the database is usable.
My question/concern is that I need to support multiple instances on the DR server, all of which will be recovering from the production site (they are basically additional test instances). Rather than expdp/impdp I'd like to use my above process as it is much faster than an import.
Can I do it? Only 1 instance on the DR server can have the same SID as the production server. Also, the file system paths are different for all but 1 of the instances on the DR server.