- Kubectl create now supports creating ingress objects. ([#94327](https://github.com/kubernetes/kubernetes/pull/94327), [@rikatz](https://github.com/rikatz)) [SIG CLI and Network]
- Kubectl rollout history sts/sts-name --revision=some-revision will start showing the detailed view of  the sts on that specified revision ([#86506](https://github.com/kubernetes/kubernetes/pull/86506), [@dineshba](https://github.com/dineshba)) [SIG CLI]

     * @return true is supplied scope is an inclusive sub-scope of current one.
     */
    public boolean encloses(ArtifactScopeEnum scope) {
        final ArtifactScopeEnum s = checkScope(scope);

        // system scope is historic only - and simple
        if (id == system.id) {
            return scope.id == system.id;
        }

        for (ArtifactScopeEnum[][] set : COMPLIANCY_SETS) {
            if (id == set[0][0].id) {

- Fixed bug in `kubectl rollout history` where only the latest revision was displayed when a specific revision was requested and an output format was specified. ([#111093](https://github.com/kubernetes/kubernetes/pull/111093), [@brianpursley](https://github.com/brianpursley))

¡No! Esa no es la moraleja de la historia.

La concurrencia es diferente del paralelismo. Y es mejor en escenarios **específicos** que implican mucha espera. Debido a eso, generalmente es mucho mejor que el paralelismo para el desarrollo de aplicaciones web. Pero no para todo.

Así que, para equilibrar eso, imagina la siguiente historia corta:

> Tienes que limpiar una casa grande y sucia.

package okhttp3.internal.http2.hpackjson

import okhttp3.internal.http2.Header
import okio.ByteString

/**
 * Representation of an individual case (set of headers and wire format). There are many cases for a
 * single story.  This class is used reflectively with Moshi to parse stories.
 */
data class Case(
  val seqno: Int = 0,
  val wire: ByteString? = null,
  val headers: List<Map<String, String>>,
) : Cloneable {

Nope! That's not the moral of the story.

Concurrency is different than parallelism. And it is better on **specific** scenarios that involve a lot of waiting. Because of that, it generally is a lot better than parallelism for web application development. But not for everything.

So, to balance that out, imagine the following short story:

> You have to clean a big, dirty house.

- Fixed a panic issue related to kubectl revision history kubernetes/kubectl#1724 ([#131496](https://github.com/kubernetes/kubernetes/pull/131496), [@tahacodes](https://github.com/tahacodes)) [SIG CLI]

* kubeadm: Implement a `--dry-run` mode and flag for `kubeadm` ([#51122](https://github.com/kubernetes/kubernetes/pull/51122), [@luxas](https://github.com/luxas))
* kubectl rollout `history`, `status`, and `undo` subcommands now support StatefulSets. ([#49674](https://github.com/kubernetes/kubernetes/pull/49674), [@crimsonfaith91](https://github.com/crimsonfaith91))

Mais les détails entre la **concurrence** et le **parallélisme** diffèrent sur de nombreux points.

Pour expliquer la différence, voici une histoire de burgers :

#### Burgers concurrents

Vous amenez votre crush 😍 dans votre fast food 🍔 favori, et faites la queue pendant que le serveur 💁 prend les commandes des personnes devant vous.

		}(i)
	}
	wg.Wait()
}

func TestEmptyReaderConcurrent(t *testing.T) {
	// Test for the race detector, to verify a Read that doesn't yield any bytes
	// is okay to use from multiple goroutines. This was our historic behavior.
	// See golang.org/issue/7856
	r := NewReader([]byte{})
	var wg sync.WaitGroup
	for i := 0; i < 5; i++ {
		wg.Add(2)
		go func() {
			defer wg.Done()
			var buf [1]byte
			r.Read(buf[:])