* Um "Init Container" no Kubernetes que roda antes do seu app container
* Um script bash que roda os passos anteriores e então inicia seu aplicativo
    * Você ainda precisaria de uma maneira de iniciar/reiniciar *aquele* script bash, detectar erros, etc.

/// tip | Dica

Darei exemplos mais concretos de como fazer isso com contêineres em um capítulo futuro: [FastAPI em contêineres - Docker](docker.md){.internal-link target=_blank}.

Sin embargo, puede haber casos en los que escribamos algún código que **colapse toda la aplicación** haciendo que Uvicorn y Python colapsen. 💥

Y aún así, probablemente no querrías que la aplicación quede muerta porque hubo un error en un lugar, probablemente querrás que **siga ejecutándose** al menos para las *path operations* que no estén rotas.

### Reiniciar Después del Colapso

github.com/eapache/go-xerial-snappy v0.0.0-20230731223053-c322873962e3/go.mod h1:YvSRo5mw33fLEx1+DlK6L2VV43tJt5Eyel9n9XBcR+0=
github.com/eapache/queue v1.1.0 h1:YOEu7KNc61ntiQlcEeUIoDTJ2o8mQznoNvUhiigpIqc=
github.com/eapache/queue v1.1.0/go.mod h1:6eCeP0CKFpHLu8blIFXhExK/dRa7WDZfr6jVFPTqq+I=
github.com/eclipse/paho.mqtt.golang v1.5.0 h1:EH+bUVJNgttidWFkLLVKaQPGmkTUfQQqjOsyvMGvD6o=

        .socket(peer.openSocket().asBufferedSocket(), "peer")
        .pushObserver(IGNORE)
        .build()
    connection.start(sendConnectionPreface = false)
    val queues = taskRunner.activeQueues()
    assertThat(queues).hasSize(1)
  }

  private fun data(byteCount: Int): Buffer = Buffer().write(ByteArray(byteCount))

  private fun assertStreamData(
    expected: String?,
    source: Source?,

- Sometimes, the scheduler incorrectly placed a pod in the "unschedulable" queue instead of the "backoff" queue. This happened when some plugin previously declared the pod as "unschedulable" and then in a later attempt encounters some other error. Scheduling of that pod then got delayed by up to five minutes, after which periodic flushing moved the pod back into the "active" queue. ([#120334](https://github.com/kubernetes/kubernetes/pull/120334), [@pohly](https://github.com/pohly))...

 *  Fix: Handle multiple 1xx responses.

 *  Fix: Address a performance bug in our internal task runner. We had a race condition that could
    result in it OkHttp starting a thread for each queued task, even when a single thread could run
    all of them.

 *  Fix: Address a performance bug in `MultipartReader`. We were scanning the entire input stream

- Implement the scheduling framework with `Reserve`, `Prebind`, `Permit`, `Post-bind`, `Queue sort` and `Unreserve` extension points.
([#77567](https://github.com/kubernetes/kubernetes/pull/77567), [@wgliang](https://github.com/wgliang))

          "rules": []
        },
        "brazilian_stop": {
          "type":       "stop",

          "rules": []
        },
        "brazilian_stop": {
          "type":       "stop",

- The scheduling queue didn't notice any extenders' failures, it could miss some cluster events,
  and it could end up Pods rejected by Extenders stuck in unschedulable pod pool in 5min in the worst-case scenario.