}

                    response.notify();
                }
            }
        } catch (final SocketTimeoutException ste) {
            log.trace("Socket timeout", ste);
        } catch (final Exception ex) {
            log.warn("Uncaught exception in NameServiceClient", ex);
        } finally {
            tryClose();
        }
    }

Falarei mais sobre imagens de contêiner, Docker, Kubernetes, etc. em um capítulo futuro: [FastAPI em contêineres - Docker](docker.md).

///

## Etapas anteriores antes de começar { #previous-steps-before-starting }

Há muitos casos em que você deseja executar algumas etapas **antes de iniciar** sua aplicação.

Por exemplo, você pode querer executar **migrações de banco de dados**.

With a response model, FastAPI will use Pydantic to serialize the data to JSON, without using intermediate steps, like converting it with `jsonable_encoder`, which would happen in any other case.

          // can actually be done anyway with the supposed 8+ exbibytes of data the source is
          // claiming to have if we get here, so just stop.
          return Optional.of(Long.MAX_VALUE);
        }
      }
      return Optional.of(result);
    }

    @Override
    public long size() throws IOException {
      long result = 0L;

Je vous en dirai plus sur les images de conteneurs, Docker, Kubernetes, etc. dans un chapitre à venir : [FastAPI dans des conteneurs - Docker](docker.md).

///

## Étapes préalables avant de démarrer { #previous-steps-before-starting }

Il existe de nombreux cas où vous souhaitez effectuer certaines étapes **avant de démarrer** votre application.

Par exemple, vous pourriez vouloir exécuter des **migrations de base de données**.

	go func() {
		// Add jitter to the update time so multiple sets don't sync up.
		updateTime := 30*time.Second + time.Duration(float64(10*time.Second)*rand.Float64())
		t := time.NewTicker(updateTime)
		defer t.Stop()
		defer saverWg.Done()
		var lastSave time.Time

		for {
			select {
			case <-t.C:
				if cache.Info.LastUpdate.Equal(lastSave) {
					continue
				}

* Datenbankverbindungen teilen.
* Sicherheit, Authentifizierung, Rollenanforderungen, usw. durchsetzen.
* Und viele andere Dinge ...

All dies, während Sie Codeverdoppelung minimieren.

## Erste Schritte { #first-steps }

Sehen wir uns ein sehr einfaches Beispiel an. Es ist so einfach, dass es vorerst nicht sehr nützlich ist.

Aber so können wir uns besser auf die Funktionsweise des **Dependency Injection** Systems konzentrieren.

labels.dict_mapping_details=Mapping Dictionary Details
labels.dict_protwords_details=Protected Words Dictionary Details
labels.dict_stemmeroverride_details=Stemmer Override Dictionary Details
labels.dict_stopwords_details=Stop Words Dictionary Details
labels.dict_synonym_details=Synonym Dictionary Details
labels.duplicate_host_details=Duplicate Host Details
labels.elevate_word_details=Elevate Word Details

     * #COMPLETED}, {@link #CANCELLED}, or {@link #INTERRUPTED}
     *
     * <p>To avoid races between threads doing release and acquire, we transition to the final state
     * in two steps. One thread will successfully CAS from RUNNING to COMPLETING, that thread will
     * then set the result of the computation, and only then transition to COMPLETED, CANCELLED, or
     * INTERRUPTED.
     *

fi

sleep 10

val=$(./mc tag list minio1/newbucket --json | jq -r .tagset | jq -r .key)
if [ "${val}" != "val1" ]; then
	echo "expected bucket tag to have replicated, exiting..."
	exit_1
fi
# stop minio1
kill -9 ${site1_pid}
# Update tag on minio2/newbucket when minio1 is down
./mc tag set minio2/newbucket "key=val2"
# create a new bucket on minio2. This should replicate to minio1 after it comes online.