A production RDS for PostgreSQL database must fail over automatically to a standby within minutes if the primary DB instance or its AZ becomes unavailable. Which configuration provides this?
A.A read replica in a second Availability Zone
Read replicas are asynchronous and must be manually promoted; they are for read scaling, not automatic failover.
B.RDS Multi-AZ deployment with synchronous standby replication
Correct. Multi-AZ synchronous standby gives automatic failover.
C.Nightly snapshots copied to another Region
Cross-Region snapshots are a DR technique with much higher RTO/RPO, not AZ failover.
D.A larger RDS instance class
Instance size affects capacity, not availability.
Why this answer
RDS Multi-AZ keeps a synchronous standby replica in a different AZ and performs automatic failover with an updated DNS endpoint — typically in 60–120 seconds — without application connection-string changes.
Correct answer: B · Last reviewed 2026-10-25
Official AWS references
Practice this interactively
Answer without seeing the solution first in the High Availability practice session.
Practice High Availability