Commit Graph

39162 Commits

Author SHA1 Message Date
Kubernetes Submit Queue
c640eeb841 Merge pull request #33260 from svanharmelen/b-cloudstack-loadbalancer
Automatic merge from submit-queue

cloudprovider/cloudstack: Fix a bug where we assume IP addresses instead of a hostnames

Because of how our test environment was setup, we didn’t notice that we were assuming the load balancer hosts list to always be IP addresses, while they actually are hostnames.

So without this PR, the load balancer code will not work as expected as it will not be able to find the nodes that need to be load balanced.

Also updated some comments and added a check to prevent trying to release a public IP if we don’t have one.
2016-11-08 21:36:16 -08:00
Kubernetes Submit Queue
ee029d9f3f Merge pull request #33850 from ymqytw/add_e2e_test_for_kubectl_in_pod
Automatic merge from submit-queue

add e2e test for kubectl in a Pod

Add a e2e test to make sure kubectl can talk to the api server when it is mounted in a pod.
Fixes: #33138
2016-11-08 21:00:53 -08:00
Kubernetes Submit Queue
b600533794 Merge pull request #36423 from Random-Liu/support-root-nobody
Automatic merge from submit-queue

CRI: Support string user name.

https://github.com/kubernetes/kubernetes/pull/33239 and https://github.com/kubernetes/kubernetes/pull/34811 combined together broke the cri e2e test. https://k8s-testgrid.appspot.com/google-gce#gci-gce-cri

The reason is that:
1) In dockershim and dockertools, we assume that `Image.Config.User` should be an integer. However, sometimes when user build the image with `USER nobody:nobody` or `USER root:root`, the field will become `nobody:nobody` and `root:root`. This makes dockershim to always return error.
2) The new kube-dns-autoscaler image is using `USER nobody:nobody`. (See https://github.com/kubernetes-incubator/cluster-proportional-autoscaler/blob/master/Dockerfile.in#L21)

This doesn't break the normal e2e test, because in dockertools [we only inspect image uid if `RunAsNonRoot` is set](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/dockertools/docker_manager.go#L2333-L2338), which is just a coincidence. However, in kuberuntime, [we always inspect image uid first](https://github.com/kubernetes/kubernetes/blob/master/pkg/kubelet/kuberuntime/kuberuntime_container.go#L141).

This PR adds literal `root` and `nobody` support. One problem is that `nobody` is not quite the same in different OS distros. Usually it should be `65534`, but some os distro doesn't follow that. For example, Fedora is using `99`. (See https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/Q5GCKZ7Q7PAUQW66EV7IBJGSRJWYXBBH/?sort=date)

Possible solution:
* Option 1: ~~Just use `65534`. This is fine because currently we only need to know whether the user is root or not.~~ Actually, we need to pass the user id to runtime when creating a container.
* Option 2: Return the uid as string in CRI, and let kuberuntime handle the string directly.

This PR is using option 1.

@yujuhong @feiskyer 
/cc @kubernetes/sig-node
/cc @MrHohn
2016-11-08 20:24:31 -08:00
nikhiljindal
d26c438a72 autogenerated bazel and test owner changes 2016-11-08 20:05:04 -08:00
nikhiljindal
1802400700 Updating deployment controller to support cascading deletion 2016-11-08 20:05:04 -08:00
nikhiljindal
54c2fbbfdf Updating federation replicaset controller to support cascading deletion 2016-11-08 20:05:04 -08:00
Kubernetes Submit Queue
6983262914 Merge pull request #36267 from vishh/gci-mounter-scope
Automatic merge from submit-queue

Make GCI nodes mount non tmpfs, ext* & bind mounts using an external mounter 

This PR downloads the stage1 & gci-mounter ACIs as part of cluster bring up instead of downloading them dynamically from gcr.io, which was the cause for #36206.

I have also optimized the containerized mounter to pre-load the mounter image once to avoid fetch latency while using it.

Original PR which got reverted: https://github.com/kubernetes/kubernetes/pull/35821

```release-note
GCI nodes use an external mounter script to mount NFS & GlusterFS storage volumes
```

@mtaufen Node e2e is not re-enabled in this PR.

cc @jingxu97
2016-11-08 19:46:32 -08:00
Kubernetes Submit Queue
5ccc8b1091 Merge pull request #36463 from nikhiljindal/debugSecretTest
Automatic merge from submit-queue

Fixing federation secret controller unit test flakiness

Fixes https://github.com/kubernetes/kubernetes/issues/36422

Adding a wait for the secret to be updated in the store to fix flakiness.
It was failing ~once in 3 to 5 runs before this change. I now have had 30 local runs without a failure.

cc @kubernetes/sig-cluster-federation @mwielgus
2016-11-08 18:02:16 -08:00
Random-Liu
021ff77028 Fix the description of MaxContainers kubelet flag. 2016-11-08 17:52:40 -08:00
Kubernetes Submit Queue
1e9344e9cc Merge pull request #36418 from jimmycuadra/fix-36323
Automatic merge from submit-queue

Use the new name for cassandra-statefulset.yaml in e2e tests.
2016-11-08 16:41:16 -08:00
Random-Liu
99ee3f4b76 Add non-numeric user name support. 2016-11-08 16:07:29 -08:00
Kubernetes Submit Queue
4b5667e177 Merge pull request #36211 from brendandburns/cordon
Automatic merge from submit-queue

Add retry to node scheduability marking.

Fixes https://github.com/kubernetes/kubernetes/issues/32391

@justinsb @janetkuo @kubernetes/sig-cli
2016-11-08 16:05:53 -08:00
Kubernetes Submit Queue
9761442b19 Merge pull request #36040 from bruceauyeung/add-master-address-desc-into-kubeadm-join-and-some-validations
Automatic merge from submit-queue

add master address into kubeadm join help message and some validations

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

1, add master address into kubeadm join help message. looks like :

>Usage:
>  kubeadm join <master address> [flags]

2, when user provides more than one master address, return an error.

3, since `kubeadm join` not only support ip addresses but also host names or domain names, so i delete the word `ip` from error message `must specify master ip address (see --help)`


Signed-off-by: bruceauyeung <ouyang.qinhua@zte.com.cn>
2016-11-08 14:52:09 -08:00
nikhiljindal
8ca1b3092f Fixing federation secret controller unit test flakiness 2016-11-08 14:08:59 -08:00
Kubernetes Submit Queue
c8a572310f Merge pull request #36459 from ixdy/fix-source-tarball
Automatic merge from submit-queue

Fix find pattern in generating kubernetes src tarball

#36407 wasn't supposed to merge. This fixes a small issue in it.
2016-11-08 14:00:06 -08:00
Kubernetes Submit Queue
860cae0933 Merge pull request #35488 from dixudx/keystone-ca-cert
Automatic merge from submit-queue

specify custom ca file to verify the keystone server

<!--  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**:

Sometimes the keystone server's certificate is self-signed, mainly used for internal development, testing and etc.

For this kind of ca, we need a way to verify the keystone server.

Otherwise, below error will occur.

> x509: certificate signed by unknown authority

This patch provide a way to pass in a ca file to verify the keystone server when starting `kube-apiserver`.

**Which issue this PR fixes** : fixes #22695, #24984

**Special notes for your reviewer**:

**Release note**:

<!--  Steps to write your release note:
1. Use the release-note-* labels to set the release note state (if you have access) 
2. Enter your extended release note in the below block; leaving it blank means using the PR title as the release note. If no release note is required, just write `NONE`. 
-->

``` release-note
```
2016-11-08 13:13:00 -08:00
Jeff Grafton
cd26a7fc5d Fix find pattern in generating kubernetes src tarball 2016-11-08 12:40:32 -08:00
Kubernetes Submit Queue
03348bb811 Merge pull request #36407 from ixdy/fix-source-tarball
Automatic merge from submit-queue

Make kubernetes-src.tar.gz contain source code again

**What this PR does / why we need it**: This PR fixes `kubernetes-src.tar.gz` by reviving the logic for `kube-source.tar.gz` from before #30787.

**Which issue this PR fixes**: fixes #36403

@saad-ali @jbeda @zmerlynn
2016-11-08 12:34:31 -08:00
Vishnu kannan
5ad53bbc86 update generated files
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-11-08 12:13:50 -08:00
Vishnu kannan
773ad9be29 Make gci mounter pre-fetch mounter image to reduce startup latency during runtime
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-11-08 12:13:49 -08:00
Jing Xu
d07396f7c7 Update configure.sh
Update the gci-mounter sha1 number
2016-11-08 12:13:49 -08:00
Jeff Grafton
8b878b0495 Make kubernetes-src.tar.gz contain source code again 2016-11-08 11:58:20 -08:00
Davanum Srinivas
cf9e9505f3 Fix build break
Problem introduced in #31996

Fixes #36454
2016-11-08 14:23:33 -05:00
Vishnu kannan
0562386385 re-enable node e2e for GCI mounter
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-11-08 11:09:13 -08:00
Vishnu kannan
77218d361b Use a local file for rkt stage1 and gci-mounter docker image.
Added a make rule `make upload` to audit and automate release artifact
uploads to GCS.

Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-11-08 11:09:13 -08:00
Vishnu kannan
dd8ec911f3 Revert "Revert "Merge pull request #35821 from vishh/gci-mounter-scope""
This reverts commit 402116aed4.
2016-11-08 11:09:10 -08:00
Kubernetes Submit Queue
2e4e932391 Merge pull request #36413 from Random-Liu/extend-node-e2e-timeout
Automatic merge from submit-queue

Node E2E: Extend the default ci node e2e test timeout to 1h.

With more and more test added into node e2e, 45m seems to be not enough now.
I saw sometimes the test takes 40+m:
* https://storage.googleapis.com/kubernetes-jenkins/logs/kubelet-gce-e2e-ci/10942/build-log.txt (44m4.917870119s)
* https://storage.googleapis.com/kubernetes-jenkins/logs/kubelet-gce-e2e-ci/10965/build-log.txt (40m2.37254827s)

And sometimes even timeout:
* https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/kubelet-gce-e2e-ci/10968
* https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/kubelet-gce-e2e-ci/10941

Although it's quite likely that the timeout happened because the limit is too tight, we are not 100% sure.
This PR extends the test timeout of regular ci node e2e test to 1h. Let's see whether the timeout will happen again.

@yujuhong
2016-11-08 11:08:12 -08:00
Tim St. Clair
7badc1d226
Use indirect streaming path for dockershim & remote CRI runtime 2016-11-08 10:58:38 -08:00
Tim St. Clair
0f028ff660
Remove legacy dockershim streaming 2016-11-08 10:58:38 -08:00
Kubernetes Submit Queue
31fbb771a2 Merge pull request #31996 from mtaufen/detect-swap-enabled
Automatic merge from submit-queue

Fail kubelet creation if swap enabled and configured with memory eviction thresholds

/cc @vishh @derekwaynecarr

**Release note**:

```
This adds an opt-in Kubelet flag (--fail-swap-on) that causes the Kubelet to fail to start if swap is enabled on the node. This is a temporary opt-in, and the Kubelet will by default fail with swap enabled starting in 1.6.0. The KubeletConfiguration equivalent to the flag is FailSwapOn.
```
2016-11-08 10:31:41 -08:00
Yu-Ju Hong
e105eec9c9 Disable keep-alive for SSH tunnel health checks
We don't reuse the http client. Disabling the keep-alive to properly close the
connections.
2016-11-08 10:03:43 -08:00
Harry Zhang
fad1990eaa Fixe verify bazel
Remove rootfs and chroot in scripts
2016-11-08 13:01:28 -05:00
Harry Zhang
64c8d3ad3d Add e2e node test for log path
Update to use pod to check log file
2016-11-08 13:01:25 -05:00
Kubernetes Submit Queue
0271eac28f Merge pull request #35765 from mfanjie/master
Automatic merge from submit-queue

register default functions for federation/apis/core

fix issue #35667 and all related e2e test fail on federation.
2016-11-08 09:53:14 -08:00
Miao Luo
20b9fc6905 Photon Controller support: Address github code review comments. 2016-11-08 09:37:20 -08:00
Miao Luo
496d229278 Fix verification error due to API version update. 2016-11-08 09:37:20 -08:00
Miao Luo
96f88f6e32 Autogenerated files update due to new API changes for photon volume 2016-11-08 09:37:20 -08:00
Miao Luo
a7fc0e4698 Photon Controller platform vendor code 2016-11-08 09:36:16 -08:00
Miao Luo
b22ccc6780 Support persistent volume on Photon Controller platform
1. Enable Photon Controller as cloud provider
2. Support Photon persistent disk as volume source/persistent volume
source
2016-11-08 09:36:16 -08:00
Michael Taufen
0c6c622434 Fail kubelet creation if swap enabled
Provides an opt-in flag, --experimental-fail-swap-on (and corresponding
KubeletConfiguration value, ExperimentalFailSwapOn), which is false by default.
2016-11-08 08:39:31 -08:00
Marcin
ada53fab42 Autogenerated stuff for PDB status observed generation 2016-11-08 17:06:18 +01:00
Marcin
8e2347370e Add observedGeneration to PodDisruptionBudgetStatus 2016-11-08 17:06:17 +01:00
Maciej Szulik
a4caa51056 Updated test_owners.csv 2016-11-08 16:28:46 +01:00
Kubernetes Submit Queue
c41c24fbf3 Merge pull request #36443 from Crassirostris/fix-fluentd-gcp-image
Automatic merge from submit-queue

Fix flunetd-gcp image Dockerfile

A mistake was made during rebasing this PR: https://github.com/kubernetes/kubernetes/pull/36370

One line, earlier deleted, was restored and broke the image build.

@piosz
2016-11-08 07:18:48 -08:00
Kubernetes Submit Queue
3b31c9f019 Merge pull request #36355 from soltysh/deprecate_extensionsjob
Automatic merge from submit-queue

 Deprecate extensions/v1beta1.Jobs related stuff

This PR supersedes https://github.com/kubernetes/kubernetes/pull/33861, it's a pre-req for removing `extensions/v1beta1.Jobs` (#32763) in the next release. 

@kubernetes/kubectl @kubernetes/api-review-team ptal
@bgrant0607 @erictune @janetkuo fyi

```release-note
Deprecate extensions/v1beta1.Jobs
```
2016-11-08 07:18:40 -08:00
Kubernetes Submit Queue
610e82079d Merge pull request #36306 from hex108/fix_kubelet_admit_message
Automatic merge from submit-queue

Fix kubelet's PodAdmitResult's message

There is no test added for it since it's a straightforward change.
2016-11-08 07:18:32 -08:00
Kubernetes Submit Queue
a2bf827e18 Merge pull request #36154 from m1093782566/m109-fix-del-ns
Automatic merge from submit-queue

fix e2e delete namespace bug

<!--  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:**

Fix `CreateNamespace()` bug in `test/e2e/framework/framework.go`.

**Why:**

If we successfully create a namespace but fail to create a ServiceAccount in it, we should delete the namespace after test.

The current implement of `CreateNamespace()` int e2e test will forget to delete the namespace which we fail to create serviceAccount in it(but the namespace has been successfully created!).
2016-11-08 07:18:24 -08:00
Kubernetes Submit Queue
c61911267f Merge pull request #35616 from pospispa/85-refactor-newRecyclerFunc-from-volume-plugins
Automatic merge from submit-queue

Simplifies NFS and hostPath plugin code

Simplifies NFS and hostPath plugin code.

cc: @jsafrane
2016-11-08 07:18:16 -08:00
Mike Danese
98e6d2c1e5 autogenerated 2016-11-08 06:33:46 -08:00
Mike Danese
584689f182 implement kubectl procelain csr commands 2016-11-08 06:33:46 -08:00