Automatic merge from submit-queue

Fixed two issues of kubectl bash completion.

This patch includes the fix of the following issue:
• Correct the method invocation from "__kubectl_namespace_flag"
  to "__kubectl_override_flags"
• Support bash completion if "--namespace=xxx" style flags are
  specified in the kubectl command

Fixes #31134

Automatic merge from submit-queue

Update fluent images

continues #29883 
fix #29782

```release-note
Reduced size of fluentd images.
```

Automatic merge from submit-queue

Add validation for KUBE_USER

Malformed KUBE_USER causes error in cluster setup.

cc/ @kubernetes/goog-image 
@Q-Lee @Amey-D  Can you please review?

Automatic merge from submit-queue

etcd3 compactor: update docs

Minor fix/improvement on compactor docs.

Automatic merge from submit-queue

Dynamic volume gate

Rebased on #31140, only review last commit.  Adds a feature-gate flag for dynamic volume provisioning alpha, defaulting to enabled to avoid breaking people. Key should be removed when support for the alpha version of this is removed.

Automatic merge from submit-queue

Revert "Enable v3 Client as the default on UTs"

Reverts kubernetes/kubernetes#30890

Fix flake problems (#31262)

Automatic merge from submit-queue

Skip disk eviction test on non-supported images.

Automatic merge from submit-queue

Delete e2e-runner and related images

https://github.com/kubernetes/test-infra/pull/444 migrates these files to test-infra

Automatic merge from submit-queue

fix #30376 that broken cross-build

ref: #30376

Automatic merge from submit-queue

Update kube-dns changelog for 1.6 bump

Fixes #29712

Automatic merge from submit-queue

Refactor the hyperkube image and add various features

@cheld @zreigz @mikedanese

Refactor hyperkube, remove unnecessary packages, optimize layers, bump cni version, add new features, run kube-proxy in a daemonset

Make sure KUBE_USER and KUBE_PASSWORD contains valid chars
as soon as we read them from kube config.

Malformed KUBE_USER causes error in cluster setup.

Automatic merge from submit-queue

Node Conformance Test: Remove unnecessary binary copy

For #30122, #30174.

This PR removed unnecessary dependencies in the node e2e test framework, because we've statically linked these dependencies.

@dchen1107 @vishh 
/cc @kubernetes/sig-node @kubernetes/sig-testing

Automatic merge from submit-queue

Kubelet: implement GetNetNS for new runtime api

Kubelet: implement GetNetNS for new runtime api. 

CC @yujuhong @thockin @kubernetes/sig-node @kubernetes/sig-rktnetes

Enabled by default to avoid breaking people, since this was
turned on in 1.3.

Automatic merge from submit-queue

Add constants and documentation around AWS magic numbers

Also, bumped max IOPS/GB to 50, it changed from 30 since last time I checked.

Source: http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/EBSVolumeTypes.html

@kubernetes/sig-storage

Automatic merge from submit-queue

Separate test watch from zero and nonzero

Recently there was spike flakes for this. See #31079.

This is to split the test into two units for further digging. We need to know which one is problematic, or both.

Automatic merge from submit-queue

Fix rc scaler watch

<!--  Thanks for sending a pull request!  Here are some tips for you:
1. If this is your first time, read our contributor guidelines https://github.com/kubernetes/kubernetes/blob/master/CONTRIBUTING.md and developer guide https://github.com/kubernetes/kubernetes/blob/master/docs/devel/development.md
2. If you want *faster* PR reviews, read how: https://github.com/kubernetes/kubernetes/blob/master/docs/devel/faster_reviews.md
3. Follow the instructions for writing a release note: https://github.com/kubernetes/kubernetes/blob/master/docs/devel/pull-requests.md#release-notes
-->

**What this PR does / why we need it**:
The RC scaler uses Watch to monitor if the RC is scaled. The Watch was sent with resourceVersion=0, so the watch event it gets back starts from any point in the history, so the result is not valid.

**Which issue this PR fixes** *(optional, in `fixes #<issue number>(, #<issue_number>, ...)` format, will close that issue when PR gets merged)*: fixes #

Probably will fix #30990
Fix #30469

**Special notes for your reviewer**:

This breaks the load test because the reaper doesn't actually wait until all the pods are gone before it deletes the RC.

I haven't run the node-100 test myself to verify it fixes the problem though.

cc @deads2k @smarterclayton @lavalamp

Automatic merge from submit-queue

Explicitly set CLOUDSDK_CONFIG override early in e2e-runner

I'm guessing this is why a number of gci-on-gke builds started failing after https://github.com/kubernetes/test-infra/pull/432 was merged - we're overriding the gcloud configuration location from the default, so the activated service account no longer works. Previously we did this service activation late, but after we'd overridden the value.

I've changed it to explicitly set to a known-good value every time at the very start.

cc @vishh @maisem @spxtr @fejta

Automatic merge from submit-queue

Revert revert 30090 with fix

This reverts #31297 (which originally reverted #30090) and applies a fix to stop the fd leak that was exposed by #30090.

Automatic merge from submit-queue

Use `make release` on Jenkins to support crossbuilds

Partner PR to https://github.com/kubernetes/test-infra/pull/440 to fix https://github.com/kubernetes/test-infra/issues/393.

We probably should pull the build functionality out of e2e.go entirely, but I leave that to future work.

cc @luxas @david-mcmahon @fejta

Automatic merge from submit-queue

Fix AWS reporting "The parameter KmsKeyId requires the parameter Encrypted to be set."

- use aws.String/Int/Bool functions
- don't set the key to empty string, use nil instead

@justinsb @kubernetes/sig-storage

Automatic merge from submit-queue

Federation controller tests refactoring - move common func to utils

cc: @quinton-hoole @kubernetes/sig-cluster-federation

Automatic merge from submit-queue

Avoid sorting lists when unnecessary

I've seen ThreadSafeMap::List consuming ~30% of whole CPU usage, spending the whole time in sorting (while it is in fact completely unneded).

Previously this code used http.Get and failed to read/close resp.Body, which
prevented network connection reuse, leaking fds. Now we use http.Head
instead, because its response always has a nil Body, so we don't have to
worry about read/close.

This reverts the revert of #30090 and #31282.

Automatic merge from submit-queue

Run Rescheduler e2e on GCE only

fix #31277

Automatic merge from submit-queue

Implement ExternalName in kube-dns

Part of the ongoing saga formerly known as https://github.com/kubernetes/features/issues/33

This is the prelude (first commit) in #30931, spawn into a separate PR to allow building a new kube-dns image before e2e tests (the rest of #30931) are updated.

ExternalName allows kubedns to return CNAME records for external
services. No proxying is involved.

cc @thockin

Automatic merge from submit-queue

Fix panics in federation e2e tets

cc: @quinton-hoole @madhusudancs

Automatic merge from submit-queue

change to stderr

`fmt.Fprint(out, "You must specify the type of resource to describe. ", valid_resources)`

change this to stderr.

Automatic merge from submit-queue

skip benchmark in jenkins serial test

This PR changes jenkins-serial.properties to skip benchmark tests (with tag [Benchmark]) in jenkins serial tests. It also add more comments in run_e2e.go.