Video on DataGuard Snapshot Standby

 I have posted a new video on creating a "Snapshot Standby" database in DataGuard.

Typically such a Standby can be used for "destructive testing" like D.R. simulation with dummy transactions, application upgrades, schema changes.

If your concern is about the Physical Standby being unavailable for long note :

(1) Redo Shipping to the Standby can continue as ArchiveLogs (of course, this means the Standby will be lagging the Primary by the typical interval between Archive Log switches)

(2) You can configure 2 Standbys of which one continues to be a Physical Standby with real time redo shipping and the other is periodically used to test schema changes / application upgrades and even for frequent D.R. testing

I have previously posted a few Blog posts on scenarios like Multiple Standbys, Standbys with Lag, Destructive testing in Standbys.  See the list of posts here.

Video on DataGuard Switchover -- with RAC and Single Instance

 I've posted a demonstration of DataGuard Switchover, using RAC and Single Instance as the Primary/Standby pair.

Fundamentally, there is no difference if either or both of the databases are RAC or Single Instance.

A Switchover is a Graceful operation, with No-Data-Loss as the Primary sends the "End-Of-Redo Marker" to the Standby at which point Reversal of Roles happens.  Therefore, you can execute Switchover between the two databases (servers / data centres / sites) multiple times without loss of data.

A Failover, on the other hand, involves data loss and the erstwhile Primary does not revert to a Standby role but must be recreated / refreshed as a Standby from the new Primary.