SAA-C03 Practice Question 36: RDS Point in Time Recovery
AWS Certified Solutions Architect – Associate (SAA-C03) · Test 4 · question 6 of 10
At 14:10 an engineer runs a faulty migration against a production Amazon RDS for PostgreSQL instance, corrupting a table. The problem is noticed at 15:30. Automated backups are enabled with a seven-day retention period, and the most recent automated snapshot was taken at 03:00 that morning. The business needs the database returned to its state immediately before 14:10, losing as little data as possible.
What should the solutions architect do?
Show answer and explanation
Answer:
With automated backups enabled, RDS continuously ships transaction logs to Amazon S3, so the database can be restored to any second within the retention window, typically up to the last five minutes. Choosing a target time just before 14:10 recovers everything committed up to the moment of the bad migration and discards only the corruption, which is the smallest possible data loss. RDS always restores to a *new* instance, so the application is repointed at the new endpoint once it is available.
Option A throws away eleven hours of valid transactions. The snapshot is a valid recovery source, but point-in-time restore uses the same backups plus the logs to do far better.
Option B does not help, because replication is continuous. A read replica applies changes from the primary within seconds, so the faulty migration was replicated to it almost immediately.
Option D misunderstands Multi-AZ. The standby is a synchronous replica maintained for availability, not a delayed copy; the corrupting statement was committed to it as part of the same transaction.
Reference: Restoring a DB instance to a specified time