Upon cross-region failover, what would happen and what would be required to continue working seamlessly regarding:

• Custom defined database directories
• Default manual backup bucket
• Default object bucket
• Object storage being used for external or partitioned data
• Default credentials APIs that interact with any of the above

CodePudding user response：

After a switchover / failover to the remote standby database, credentials, object store URLs to the primary region object store (including for manual backups, external / hybrid partitioned tables) will continue to work as they did on the primary database - Assuming the primary region's object store is still up and running.

If the primary region's object storage is down / unavailable due to a disaster, and users wish to use a different region's bucket, users must update their URLs to point to that different region's bucket URLs.

Custom defined database directory paths are local to the current primary database and should continue to work after a switchover / failover without any changes necessary.

Database APIs calls in general are specific to each database (ie. which has its own OCID), and therefore after a failover / switchover, users must call database APIs with the new (remote) primary's database OCID as the “autonomousDatabaseId”.