* kubeadm: fix a case where fetching a kubernetesVersion from the internet still happened even if some commands don't need it. ([#69645](https://github.com/kubernetes/kubernetes/pull/69645), [@neolit123](https://github.com/neolit123))

- Fix kubelet to properly log when a container is started. Previously, kubelet may log that container is dead and was restarted when it was actually started for the first time. This behavior only happened on pods with initContainers and regular containers. ([#91469](https://github.com/kubernetes/kubernetes/pull/91469), [@rata](https://github.com/rata))

- 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"...

- 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"...

  <artifactId>inheritance</artifactId>
  <version>11-SNAPSHOT</version>
  <name>Model urls inheritance test child</name>

  <scm child.scm.connection.inherit.append.path="false"
       child.scm.developerConnection.inherit.append.path="false"
       child.scm.url.inherit.append.path="false">
    <connection>scm:my-scm:http://domain.org/base</connection>
    <developerConnection>scm:my-scm:https://domain.org/base/</developerConnection>

  child.project.url.inherit.append.path="false">
  <modelVersion>4.0.0</modelVersion>

  <groupId>inheritance</groupId>
  <artifactId>parent</artifactId>
  <version>11-SNAPSHOT</version>

  <name>Model urls inheritance test parent, with special merge test for mergeSite_ChildSiteUrlInheritAppendPath</name>

  <scm child.scm.connection.inherit.append.path="false"

  <description>MNG-5951 MNG-6059 child.x.y.inherit.append.path="false" for each url to avoid automatic path addition when inheriting</description>

  <!-- 5 inherited urls without anything added to parent -->
  <url>http://www.apache.org/path/to/parent/</url>
  <scm child.scm.connection.inherit.append.path="false"
       child.scm.developerConnection.inherit.append.path="false"
       child.scm.url.inherit.append.path="false">

  </modules>

  <!-- 5 urls in the pom to configure for not adding path -->
  <url>http://www.apache.org/path/to/parent/</url>
  <scm child.scm.connection.inherit.append.path="false"
       child.scm.developerConnection.inherit.append.path="false"
       child.scm.url.inherit.append.path="false">
    <connection>scm:my-scm:http://domain.org/base</connection>
    <developerConnection>scm:my-scm:https://domain.org/base/</developerConnection>

- 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"...

  <scm>
    <url>https://domain.org/override</url><!-- check that this won't impact child.site.url.inherit.append.path attribute inheritance -->
  </scm>
  <distributionManagement>
    <site><!-- overriding 1 element will reset all elements, but not child.site.url.inherit.append.path attribute -->
      <name>reset</name>
    </site>
  </distributionManagement>