RAC in the workplace: what are common backup setups for lifecycle databases?

Good day, list,

We have been trying to determine what an ideal layout for database backups given the following:
- 3 sets of 2-node clusters, DEV, BCP and PROD

• Production cluster is 130 miles away from the other two
• we often clone the production environment (from a standby) to development
• we often have to manipulate archived logs for standbys and downstream capture databases
• Oracle 10.2.0.3 on RHEL4
• EMC SAN used at all locations (both DMX and Clariion)
• Currently, rman backups go to +FRA and a cooked filesystem to spool to tape

To date, ASM is not shared across DEV, BCP or PROD. I want to explore that idea, as there are extra steps to move files around when dealing with ASM. But first, I am curious what other folks are doing in these situations. The books and whitepapers that we have read mostly talk about single-environment clusters (ie, Production *or* Development, but not both). We would like to find a way that is easy to maintain; if someone has a great, practical suggestion for ASM, I am willing to listen. My other thought is using a NFS mount and sharing that across clusters. We do not have any plans to utilize Flashback database, so the idea of keeping our backups in the FRA at this point in time is purely so that each instance in a cluster can access them (especially the archive logs).

-- 
Charles Schultz

--
http://www.freelists.org/webpage/oracle-l