A company uses an Amazon Aurora DB cluster in a Multi-AZ configuration. During a planned failover of the writer instance, the database endpoints in the application are updated incorrectly. After failover, reads work but writes fail with connection errors and timeouts for several minutes. The team currently uses the instance endpoint for the writer. What should they change to improve write resilience during failovers?

Continue using the instance endpoint, but increase application retry count so the writer changes are handled more quickly.

Retries can reduce perceived impact, but the instance endpoint can become non-writer after failover. If the application continues targeting an instance endpoint that no longer accepts writes, retries will continue to fail until the code or endpoint selection is corrected.

Use a read replica endpoint for writes because it is typically stable across failovers.

Read replica (reader) endpoints are intended for read workloads. They can route to non-writer instances, and those endpoints should not be used for writes because they may reject write operations.

Disable Multi-AZ failover so the writer instance never changes and writes remain consistent.

Disabling failover removes an important resiliency mechanism and increases downtime risk. It improves endpoint stability only by trading away availability during instance or AZ issues.

What does this SAA-C03 question test?

Static NAT maps one inside address to one outside address.

What is the correct answer to this question?

The correct answer is: Use the Aurora cluster writer endpoint for all write operations. — In Aurora, instance endpoints are tied to a specific instance. After failover, the writer role moves to a different instance, so an application that connects to the old writer instance endpoint may start experiencing write failures and timeouts. For resilient write connectivity, the application should use the Aurora cluster writer endpoint, which is meant for write traffic and automatically points to the current writer instance after failover. This eliminates brittle manual endpoint updates and improves write availability. Increasing retries does not fix the underlying routing problem when writes are directed to an instance endpoint that is no longer the writer. Read replica endpoints are not guaranteed to route to the writer and are not intended for writes. Disabling failover increases the likelihood of prolonged write outages during failures.

What should I do if I get this SAA-C03 question wrong?

Then try more questions from the same exam bank and focus on understanding why the wrong options are tempting.