Exercício prático · Cloud / AWS

Practice lab: Process the same event twice safely

do caderno de estudos · atualizado em 2 de outubro de 2026
Para quem vai encarar entrevista de Cloud / AWS2 min

Tente resolver a tarefa antes de consultar os critérios. Compare o resultado e explique o que mudou no seu raciocínio.

Tarefa e critérios de verificação

Task, 15 minutes: design S3 upload → SQS → thumbnail worker. An object may be overwritten, a queue message may be delivered twice, and one corrupt image repeatedly fails. State your work-duration and resource assumptions, then use them to justify a worker runtime. Define the work identity, retry/dead-letter policy, and required worker permissions.

Check your work:

  • The logical work key includes object identity, object version or an equivalent immutable input identity, and transformation version. A later overwrite is not mistaken for an old completed job.
  • A completed retry produces the same intended thumbnail result without creating duplicate business records.
  • A transient failure can retry after visibility timeout. Poison messages reach a configured dead-letter queue after the allowed attempts.
  • The worker role can read the required source and write the required destination without broad account permissions.

Follow-up: describe how to inspect and replay a failed message in a test environment after fixing the cause. Increasing retries alone does not repair a corrupt input.

No app, a tarefa fica visível e você pode revelar os critérios depois da tentativa. Use as notas para registrar sua solução.

Praticar este exercício no app