val server = serverListener.assertOpen()

    // Initiate a close on the client, which will schedule a hard cancel in 500 ms.
    val closeAtNanos = System.nanoTime()
    webSocket.close(1000, "goodbye", 500L)
    serverListener.assertClosing(1000, "goodbye")

    // Confirm that the hard cancel occurred after 500 ms.
    clientListener.assertFailure()
    val elapsedUntilFailure = System.nanoTime() - closeAtNanos

			testWrite{"hello\n", 6, nil},
			testClose{nil},
		},
	}, {
		// This file was produced using GNU tar v1.26:
		//	echo "Slartibartfast" > file.txt
		//	ln file.txt hard.txt
		//	tar -b 1 --format=ustar -c -f hardlink.tar file.txt hard.txt
		file: "testdata/hardlink.tar",
		tests: []testFnc{
			testHeader{Header{
				Typeflag: TypeReg,
				Name:     "file.txt",
				Size:     15,
				Mode:     0644,

Bit Rot, also known as data rot or silent data corruption is a data loss issue faced by disk drives today. Data on the drive may silently get corrupted without signaling an error has occurred, making bit rot more dangerous than a permanent hard drive failure.

MinIO's erasure coded backend uses high speed [HighwayHash](https://github.com/minio/highwayhash) checksums to protect against Bit Rot.

## How are drives used for Erasure Code?

            }
        }
    }

    protected synchronized void doDisconnect(final boolean hard) throws IOException {
        doDisconnect(hard, false);
    }

    @Override
    protected synchronized boolean doDisconnect(final boolean hard, final boolean inUse) throws IOException {
        final ListIterator<SmbSessionImpl> iter = this.sessions.listIterator();

	return tillOffset
}

// erasureSelfTest performs a self-test to ensure that erasure
// algorithms compute expected erasure codes. If any algorithm
// produces an incorrect value it fails with a hard error.
//
// erasureSelfTest tries to catch any issue in the erasure implementation
// early instead of silently corrupting data.
func erasureSelfTest() {
	// Approx runtime ~1ms
	var testConfigs [][2]uint8

	"io"
	"sync"
	"time"

	"github.com/minio/madmin-go/v3"
)

// naughtyDisk wraps a POSIX disk and returns programmed errors
// specified by the developer. The purpose is to simulate errors
// that are hard to simulate in practice like DiskNotFound.
// Programmed errors are stored in errors field.
type naughtyDisk struct {
	// The real disk
	disk StorageAPI
	// Programmed errors: API call number => error to return

## Campos de lista

Puedes definir un atributo como un subtipo. Por ejemplo, una `list` en Python:

{* ../../docs_src/body_nested_models/tutorial001_py310.py hl[12] *}

Esto hará que `tags` sea una lista, aunque no declare el tipo de los elementos de la lista.

## Campos de lista con parámetro de tipo

Pero Python tiene una forma específica de declarar listas con tipos internos, o "parámetros de tipo":

   There used to be two different mechanisms - one for regular S3 objects and one for IAM data.
- Reduced server startup time. For IAM encryption with the root credentials, MinIO had
   to use a memory-hard function (Argon2) that (on purpose) consumes a lot of memory and CPU.
   The new KMS-based approach can use a key derivation function that is orders of magnitudes
   cheaper w.r.t. memory and CPU.

* tenerlos en un orden diferente
* no usar `Annotated`

...Python tiene una sintaxis especial para eso.

Pasa `*`, como el primer parámetro de la función.

Luego tu API (imaginemos):

* Enviará la factura a algún cliente del desarrollador externo.
* Recogerá el dinero.
* Enviará una notificación de vuelta al usuario de la API (el desarrollador externo).
    * Esto se hará enviando un request POST (desde *tu API*) a alguna *API externa* proporcionada por ese desarrollador externo (este es el "callback").

## La aplicación normal de **FastAPI**