Automatic merge from submit-queue

Add support for 3rd party objects to kubectl

@deads2k @jlowdermilk

Instructions for playing around with this:

Run an apiserver with third party resources turned on (`--runtime-config=extensions/v1beta1=true,extensions/v1beta1/thirdpartyresources=true`)

Then you should be able to:

```
kubectl create -f rsrc.json
```

```json
{
  "metadata": {
    "name": "foo.company.com"
  },
  "apiVersion": "extensions/v1beta1",
  "kind": "ThirdPartyResource",
  "versions": [
    {
      "apiGroup": "group",
      "name": "v1"
    },
    {
      "apiGroup": "group",
      "name": "v2"
    }
  ]
}
```

Once that is done, you should be able to:

```
curl http://<server>/apis/company.com/v1/foos
```

```
curl -X POST -d @${HOME}/foo.json http://localhost:8080/apis/company.com/v1/namespaces/default/foos
```

```json
{
  "kind": "Foo",
  "apiVersion": "company.com/v1",
  "metadata": {
    "name": "baz"
  },
  "someField": "hello world",
  "otherField": 1
}
```

After this PR, you can do:

```
kubectl create -f foo.json
```

```
kubectl get foos
```

etc.

Automatic merge from submit-queue

fix the gcfg typo

little fix in the openstack cloud provider.

Automatic merge from submit-queue

Migrate to the new conversion generator - part1

This PR contains two commits:
- few more fixes to the generator
- migration of the pkg/api/v1 to use the new generator

The second commit is big, but I reviewed the changes and they contain:
- conversions between types that we didn't even generating conversion between
- changes in how we handle maps/pointers/slices - previously we were explicitly referencing fields, now we are using "shadowing in, out" to make the code more generic
- lack of auto-generated method for ReplicationControllerSpec (because these types are different (*int vs int for Replicas) and a preexisting conversion already exists

Most of issues in the first commit (e.g. adding references to "in" and "out" for slices/maps/points) were discovered by our tests. So I'm pretty confident that this change is correct now.

Ignore .config/gcloud-staging/ in .gitignore

Stop swallowing stdout and stderr when we check for the staging bucket.

Remove unnecessary override of /etc/init.d/docker on containervm image.

Kubemark uses kubectl from the repo

Temporarily disable goroutine calls to testing.T.Logf to avoid race

Automatic merge from submit-queue

Retry github and godep operations in test-dockerized.sh

closes #21887.

Attempt to mitigate go get and godep flakes by retrying a few times inside of jenkins

Automatic merge from submit-queue

storage.Interface KV impl. of etcd v3

This is the initial implementation of #22448.

The PR consists of two parts:
- add godep of "clientv3" and "integration" (for testing)
- create new package "etcd3" under "pkg/storage/"
- implement KV methods of storage.Interface using etcd v3 APIs
  - Create, Set, Get, Delete, GetToList, List, GuaranteedUpdate

make docker-checker more robust

Automatic merge from submit-queue

Add a protobuf serializer

Provide a core protobuf serializer that can either write objects with
an envelope (a 4 byte prefix and a runtime.Object) or raw to a byte
array.

Makes a few small refactors to prepare for streaming codecs (watch)
including a new constructor for the codec factory.

@kubernetes/sig-api-machinery the bulk of this is not compiled by
default (under the 'proto' tag), but lays groundwork the enablement
here.

Provide a core protobuf serializer that can either write objects with
an envelope (a 4 byte prefix and a runtime.Object) or raw to a byte
array.

Automatic merge from submit-queue

When checking for leak look only at additional resources

This should help with "fake" leaks, when run deletes stuff that was leaked in a previous one.

cc @zmerlynn @ixdy @wojtek-t 

Automatic merge from submit-queue

Implement network plugin capabilities hook and shaping capability

Allow network plugins to declare that they handle shaping and that
Kuberenetes should not.  I've got an OpenShift PR that handles shaping in OVS but the kubelet code sends nasty pod events because it doesn't think shaping is used since --reconcile-cbr0 is not used with most network plugins.

See https://github.com/openshift/openshift-sdn/pull/266 for the OpenShift implementation.

Change kube-proxy & fluentd CPU request to 20m/80m.

Automatic merge from submit-queue

Explicit links needed for PRs and users in github source markdown.

Fix some other formatting:
- Reverse ordered the included releases for 1.2.0 (newest to oldest).
- Use the same Downloads format for all releases so far (in this file).

Automatic merge from submit-queue

Fixed typo.

Corrected log message.

Update guestbook examples; replace RC concepts with Deployment

Automatic merge from submit-queue

Release Notes proposal

Get the conversation started.
cc @eparis @vishh @davidopp @thockin @dchen1107 
Ref #23070

Automatic merge from submit-queue

kubectl: enhance podtemplate describer

@kubernetes/kubectl 

@smarterclayton for the dc decsriber

Fix cluster names to be numeric in upgrade jobs

This implements Get(), Create(), Delete(), GetToList(),
List(), GuaranteedUpdate().

Decrease verbosity of namespace controller trace logging

Update our GCP metadata dependency to include a recent flakiness improvement

Allow network plugins to declare that they handle shaping and that
Kuberenetes should not.  Will be first used by openshift-sdn which
handles shaping through OVS, but this triggers a warning when
kubelet notices the bandwidth annotations.

Let Noop handle common functions.

Automatic merge from submit-queue

Update conformance test policy

Addresses #14913

/cc @kubernetes/sig-testing

Automatic merge from submit-queue

rkt: bump rkt version to 1.2.1

Upon bumping the rkt version, `--hostname` is supported. Also we now gets the configs from the rkt api service, so `stage1-image` is deprecated. 

cc @yujuhong @Random-Liu 

Automatic merge from submit-queue

allow kubectl cmds to process files recursively, when given a dir

This PR accomplishes two things:

1. It creates a `--recursive` flag for use with certain `kubectl` commands that _currently_ do not process files beyond their first level of children, as seen in the issue https://github.com/kubernetes/kubernetes/issues/19767
2. It enables the ability to use the `--recursive` flag in the `kubectl` commands that currently _only_ support processing a directory up to its first level of children.