margin-bottom: 0;\n  @include border-radius(var(--#{$prefix}list-group-border-radius));\n}\n\n.list-group-numbered {\n  list-style-type: none;\n  counter-reset: section;\n\n  > .list-group-item::before {\n    // Increments only this instance of the section counter\n    content: counters(section, \".\") \". \";\n    counter-increment: section;\n  }\n}\n\n// Interactive list items\n//\n// Use anchor or button elements instead of `li`s or `div`s to create interactive\n// list items. Includes an extra...

    <magic priority="50">
      <!-- Variable number of spaces or possibly creator tool before the code groups -->
      <!-- allow 999 (comments) before section -->
      <match type="regex" value="(999\r?\n[^\r\n]{0,64}\\s+)?0\r?\nSECTION\r?\n" offset="0:32">
        <match type="regex" value="2\r?\n(?:HEADER|ENTITIES)\r?\n" offset="12:60"/>
      </match>
    </magic>
  </mime-type>

- Fixed the issue where logging output of kube-scheduler configuration files included line breaks and escape characters. The output also attempted to output the configuration file in one section without showing the user a more readable format. ([#106228](https://github.com/kubernetes/kubernetes/pull/106228), [@sanchayanghosh](https://github.com/sanchayanghosh)) [SIG Scheduling]

- Commands like `kubectl apply` now return errors if schema-invalid annotations are specified, rather than silently dropping the entire annotations section. ([#83552](https://github.com/kubernetes/kubernetes/pull/83552), [@liggitt](https://github.com/liggitt))

# Release 2.21.0

## TensorFlow

<INSERT SMALL BLURB ABOUT RELEASE FOCUS AREA AND POTENTIAL TOOLCHAIN CHANGES>

### Breaking Changes

* <DOCUMENT BREAKING CHANGES HERE>
* <THIS SECTION SHOULD CONTAIN API, ABI AND BEHAVIORAL BREAKING CHANGES>

### Known Caveats

* <CAVEATS REGARDING THE RELEASE (BUT NOT BREAKING CHANGES).>
* <ADDING/BUMPING DEPENDENCIES SHOULD GO HERE>
* <KNOWN LACK OF SUPPORT ON SOME PLATFORM, SHOULD GO HERE>

## Urgent Upgrade Notes

### (No, really, you MUST read this before you upgrade)

#### Cluster Lifecycle

- Container images tar files for `amd64` will now contain the architecture in the RepoTags manifest.json section.
  If you are using docker manifests there are not visible changes. ([#80266](https://github.com/kubernetes/kubernetes/pull/80266), [@javier-b-perez](https://github.com/javier-b-perez))

  
  Changes in Pull Request:
  
    1. Add properties `networkResourceTenantID` and `networkResourceSubscriptionID` in cloud provider auth config section, which indicates the location of network resources.
    2. Add function `GetMultiTenantServicePrincipalToken` to fetch multi-tenant service principal token, which will be used by Azure VM/VMSS Clients in this feature.

* Fix failing load balancers in Azure ([#40405](https://github.com/kubernetes/kubernetes/pull/40405), [@codablock](https://github.com/codablock))
* Add a KUBERNETES_NODE_* section to build kubelet/kube-proxy for windows ([#38919](https://github.com/kubernetes/kubernetes/pull/38919), [@brendandburns](https://github.com/brendandburns))

- Kubeadm: add the `imagePullPolicy` field in the `nodeRegistration` section of `InitConfiguration` and `JoinConfiguration` in `v1beta3`. This allows the user to specify the image pull policy during "kubeadm init" and "kubeadm join". The value of this field must be one of `Always`, `IfNotPresent` or `Never`. The default behavior...

### SIG Network

The main milestones for SIG Network this release are the graduation of IPVS-based load balancing and CoreDNS to general availability.