What AWS service feature should an application developer use to ensure that messages which cannot be processed after a certain number of retries are not lost and can be analyzed for further investigation?
Retry policies with immediate retries
Dead-letter queues
Stateless application patterns
Scheduled scaling actions