A DBA executed a script which corrupted an Oracle DB managed by NDB.

After a DBA’s script corrupts an Oracle DB managed by NDB, the administrator needs to restore a clean copy using the latest available snapshot. The most straightforward action is to use the NDB Time Machine feature to select the most recent uncorrupted snapshot and create a new clone. This clone can serve as a restored version of the database, bypassing the corrupted data while leveraging existing snapshots for efficiency.

Option A (Fail the Oracle cluster over to a surviving RAC node) is incorrect because RAC failover is for HA, not corruption recovery via snapshots.

Option B is correct as it uses Time Machine to create a clone from a valid snapshot, addressing the corruption.

Option C (Use Time Machine to create a clone from file-level database backup) is incorrect because NDB primarily uses snapshots, not file-level backups, for cloning.

Option D (Provision a new Oracle DB and perform log catch-up) is incorrect because provisioning a new DB is more complex and unnecessary when a snapshot-based clone suffices.

This method ensures rapid recovery with minimal disruption.

References

Nutanix Database Service (NDB) User Guide, Chapter 5: Configuring Time Machines, Section: Restoring and Cloning from Snapshots

Nutanix Support & Insights, Knowledge Base Article: "Recovering from Database Corruption in NDB"

Nutanix Certified Professional - Database Automation (NCP-DB) v6.5 Blueprint, Section 5: Protect Databases Using Time Machine