Automatic merge from submit-queue

Move post processing/backward compatibility of openapi out of generic package

Post processing step was put into generic package and was the same for both k8s api server and federation api server. They have different backward compatibility list of types. This PR move that step out of generic package and put it in each server's file.

Automatic merge from submit-queue

genericapiserver: cut off more dependencies – episode 6

Follow-up of https://github.com/kubernetes/kubernetes/pull/40808

approved based on #40363

Automatic merge from submit-queue (batch tested with PRs 40556, 40720)

Emit events on 'Failed' daemon pods

Follow up #40330 @erictune @mikedanese @kargakis @lukaszo @kubernetes/sig-apps-bugs

Automatic merge from submit-queue

kubeadm: Refactoring the apiconfig and addons phases 

**What this PR does / why we need it**:

First commit: Fix the defaulting for AuthorizationMode - **Ready for review**
Second commit: Refactoring the apiconfig and addons phases in kubeadm - **work in progress, broken at the moment**

**Special notes for your reviewer**:

Please take a look at the first commit now. You can also see the direction I'm going with the addons and apiconfig phases.

For example, I'm using Go templates instead of creating native Go structs for kube-dns and kube-proxy now. The question is if I should do it for the RBAC rules as well, it might make sense there as well.

Converting the dns addon to a yaml spec makes it 100x easier to maintain when changes are made to the "upstream" DNS spec, and also more swappable. We could for instance have a configuration param for it for those who want a custom file, etc.

**Release note**:

```release-note
NONE
```

@mikedanese @pires @lukemarsden @errordeveloper @dgoodwin @liggitt @andrewrynhard

Automatic merge from submit-queue (batch tested with PRs 39169, 40719, 38954, 40808, 40689)

add details to patch conflict

Adds conflict details to the patch message so we can debug #39471.  Seems like this may have general utility since others may also wonder what they conflicted on.

@sig-api

Automatic merge from submit-queue (batch tested with PRs 39169, 40719, 38954, 40808, 40689)

genericapiserver: cut off more dependencies – episode 5

Follow-up of https://github.com/kubernetes/kubernetes/pull/40658

Left for episode 6:
- pkg/storage/etc
- pkg/storage/etc3
- pkg/storage/storagebackend/factory

approved based on #40363

Automatic merge from submit-queue (batch tested with PRs 39169, 40719, 38954, 40808, 40689)

Add StatefulSets checks at Service level

Hi!

Please let me propose some very small e2e testsuite enhancement. 

This PR removed a `TODO` about checking governing service at unit test level (which is hard) and adds this to e2e testsuite.

Thanks
Sebastian

Automatic merge from submit-queue (batch tested with PRs 39169, 40719, 38954, 40808, 40689)

Exports KUBE_TEMP for use in Vagrantfile

In #40147, the logic for setting `KUBE_TEMP` was refactored into `common.sh`. However, it was overlooked that `KUBE_TEMP` [needs to be exported for vagrant to work properly](https://github.com/kubernetes/kubernetes/pull/40147/files#diff-b19d3d93456020e2168c7f304f722969).

This PR restores the `export` so that `Vagrantfile` can use `ENV["KUBE_TEMP"]` properly.

👀 @rthallisey @shyamjvs @timothysc

Automatic merge from submit-queue (batch tested with PRs 39169, 40719, 38954, 40808, 40689)

add systemd notify

Automatic merge from submit-queue

Add websocket support for port forwarding

#32880

**Release note**:
```release-note
Port forwarding can forward over websockets or SPDY.
```

Automatic merge from submit-queue (batch tested with PRs 39217, 40624)

Allow StorageFactory to wrap encoders and decoders

Automatic merge from submit-queue

Added missing closing parens to kubectl describe node command

Closing parens was missing in the "Total limits may be over 100%" line. See below:
```
$ kubectl describe node
Name:			minikube
...
Allocated resources:
  (Total limits may be over 100 percent, i.e., overcommitted.
  CPU Requests	CPU Limits	Memory Requests	Memory Limits
...
```

```release-note
Improve kubectl describe node output by adding closing paren
```

Automatic merge from submit-queue (batch tested with PRs 37617, 40197)

dnsprovider: Add direct access to Route53 data

This methods allow us to develop advanced functionality for Route53,
before we add all the functionality to the cross-provider interface.
Use of these methods should be avoided, and adding methods to the
cross-provider interfaces should be preferred.

```release-note
NONE
```

Automatic merge from submit-queue (batch tested with PRs 37617, 40197)

fix typo in daemon

fix typo in daemon.

Automatic merge from submit-queue (batch tested with PRs 40828, 40821)

test/images/mount-tester-user: bump base image to 0.8

This PR bumps the base image used by `mount-tester-user` after 
https://github.com/kubernetes/kubernetes/pull/40613.

Automatic merge from submit-queue

Bump GCE to container-vm-v20170201

```release-note
Bump GCE ContainerVM to container-vm-v20170201 to address CVE-2016-9962.
```

- adjust ports to int32
- CRI flows the websocket ports as query params

- Do not validate ports since the protocol is unknown
  SPDY flows the ports as headers and websockets uses query params
- Only flow query params if there is at least one port query param

Automatic merge from submit-queue (batch tested with PRs 40574, 40806, 40308, 40771, 39440)

kubeadm: kube-proxy needs to know the pod subnet CIDR

**What this PR does / why we need it**: `kube-proxy` 1.5 has a new flag `cluster-cidr` that isn't specified by `kubeadm`, thus resulting in bug https://github.com/kubernetes/kubeadm/issues/102.

**Which issue this PR fixes**: fixes https://github.com/kubernetes/kubeadm/issues/102

**Special notes for your reviewer**:
/cc @luxas @dmmcquay

Automatic merge from submit-queue (batch tested with PRs 40574, 40806, 40308, 40771, 39440)

kubeadm: preflight check for incorrect FQDN

**What this PR does / why we need it**: There are a variety of system configuration errors (such as cloud-init bugs when deploying on AWS) which can cause hostname and uname -n to be wrong for a given host. This will cause kubeadm setup to fail in interesting and hard-to-figure-out ways (it doesn't fail until you start trying to set up DNS on the master, for example).

This PR adds a preflight check to test whether or not the server can reach itself using that name. This does not catch the case that the FQDN belongs to a different but valid server, but it would catch some of the cases. 

**Which issue this PR fixes** : fixes https://github.com/kubernetes/kubeadm/issues/135

**Special notes for your reviewer**: /cc @luxas 

**Release note**:
```release-note
NONE
```