Commit Graph

686 Commits (04326905b8d31fb309fd5306f3741e230b5bf4df)

Author SHA1 Message Date
deads2k 6a4d5cd7cc start the apimachinery repo 2017-01-11 09:09:48 -05:00
Kubernetes Submit Queue 3f9f7471af Merge pull request #38989 from sjenning/set-qos-field
Automatic merge from submit-queue (batch tested with PRs 39684, 39577, 38989, 39534, 39702)

Set PodStatus QOSClass field

This PR continues the work for https://github.com/kubernetes/kubernetes/pull/37968

It converts all local usage of the `qos` package class types to the new API level types (first commit) and sets the pod status QOSClass field in the at pod creation time on the API server in `PrepareForCreate` and in the kubelet in the pod status update path (second commit).  This way the pod QOS class is set even if the pod isn't scheduled yet.

Fixes #33255

@ConnorDoyle @derekwaynecarr @vishh
2017-01-10 22:24:13 -08:00
Kubernetes Submit Queue a2da4f0cac Merge pull request #39546 from dashpole/dynamic_config_eviction_hard
Automatic merge from submit-queue (batch tested with PRs 39695, 37054, 39627, 39546, 39615)

Use Dynamic Config in e2e_node inode eviction test

Alternative solution to #39249.  Similar to solution proposed by @vishh in #36828.

@Random-Liu @mtaufen
2017-01-10 18:57:26 -08:00
Jeff Grafton 19aafd291c Always --pull in docker build to ensure recent base images 2017-01-10 16:21:05 -08:00
Seth Jennings e2402b781b set qos class field in pod status 2017-01-10 16:31:52 -06:00
Seth Jennings 4c30459e49 switch from local qos types to api types 2017-01-10 10:54:30 -06:00
David Ashpole c3951a72ab use dynamic config to set eviction hard threshold 2017-01-09 15:27:12 -08:00
Jeff Grafton 20d221f75c Enable auto-generating sources rules 2017-01-05 14:14:13 -08:00
Kubernetes Submit Queue f4a8713088 Merge pull request #36229 from wojtek-t/bump_etcd_version
Automatic merge from submit-queue (batch tested with PRs 36229, 39450)

Bump etcd to 3.0.14 and switch to v3 API in etcd.

Ref #20504

**Release note**:

```release-note
Switch default etcd version to 3.0.14.
Switch default storage backend flag in apiserver to `etcd3` mode.
```
2017-01-04 17:36:06 -08:00
Mike Danese 161c391f44 autogenerated 2016-12-29 13:04:10 -08:00
Random-Liu a719a7d7e7 Do not use sudo when untar node e2e tar ball. 2016-12-21 16:28:33 -08:00
Kubernetes Submit Queue 1955ed614f Merge pull request #39074 from Random-Liu/node-e2e-set-user
Automatic merge from submit-queue

Node E2E: Set user with `--ssh-user` flag when running remote node e2e.

This PR unblocks https://github.com/kubernetes/test-infra/issues/1348.

In our test environment, we must login test instance as user `jenkins` because of the service account. Node e2e is always using the default user on the host, which works fine till now, because it is always run as `jenkins` in our test environment.

However, now we moved the test runner into a docker container, inside the container user is `root` by default, which will cause error:
```
Permission denied (publickey)
```

This PR added a flag `--ssh-user` to explicitly specify the user used to ssh into test instance. The dockerized test runner can set user to `jenkins` with this flag.

@krzyzacy  @ixdy
2016-12-21 11:21:09 -08:00
Random-Liu 10f72be5af Support set user with `--ssh-user` flag when running remote node e2e. 2016-12-21 01:54:02 -08:00
Wojciech Tyczynski 498a893fa3 Switch to etcd v3 API by default 2016-12-20 11:57:46 +01:00
Kubernetes Submit Queue 5b2823adb9 Merge pull request #38191 from sttts/sttts-move-master-options
Automatic merge from submit-queue

Move non-generic apiserver code out of the generic packages
2016-12-17 01:25:45 -08:00
Matt Liggett 69cd805532 Merge pull request #38804 from Random-Liu/disable-au
Node E2E: Disable AU in node e2e test.
2016-12-16 15:32:23 -08:00
David Ashpole 5d352439d4 test no longer fails when it fails to get the summary 2016-12-16 11:50:43 -08:00
Dr. Stefan Schimanski 7267299c3c genericapiserver: move MasterCount and service options into master 2016-12-16 17:23:43 +01:00
Random-Liu c57f2ec064 Fix report prefix for node conformance test. 2016-12-15 15:27:14 -08:00
Kubernetes Submit Queue d169d59565 Merge pull request #38788 from Random-Liu/fix-node-conformance-test
Automatic merge from submit-queue (batch tested with PRs 38788, 38821, 38829)

Node Conformance Test: Fix node conformance test.

The test suite could build on my desktop. However it is failing on jenkins.
https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/ci-kubernetes-node-kubelet-conformance/1

It turns out that `docker save $IMAGE -o $FILE` only works for docker 1.12. (My desktop is 1.12) For older version docker, we should use `docker save -o $FILE $IMAGE instead`. (Jenkins is using 1.9.1)

@timstclair Could you help me review this short PR? :)
2016-12-15 13:57:16 -08:00
Random-Liu e5efc21de6 Disable AU in node e2e test. 2016-12-15 01:33:09 -08:00
Kubernetes Submit Queue d8efc779ed Merge pull request #38154 from caesarxuchao/rename-release_1_5
Automatic merge from submit-queue (batch tested with PRs 38154, 38502)

Rename "release_1_5" clientset to just "clientset"

We used to keep multiple releases in the main repo. Now that [client-go](https://github.com/kubernetes/client-go) does the versioning, there is no need to keep releases in the main repo. This PR renames the "release_1_5" clientset to just "clientset", clientset development will be done in this directory.

@kubernetes/sig-api-machinery @deads2k 

```release-note
The main repository does not keep multiple releases of clientsets anymore. Please find previous releases at https://github.com/kubernetes/client-go
```
2016-12-14 14:21:51 -08:00
Random-Liu 06507e9378 Fix node conformance test. 2016-12-14 14:21:10 -08:00
Chao Xu 03d8820edc rename /release_1_5 to /clientset 2016-12-14 12:39:48 -08:00
Random-Liu 02e96df55c Update log level. 2016-12-13 19:08:55 -08:00
Random-Liu 54c874f2c6 Update bazel. 2016-12-13 19:08:55 -08:00
Random-Liu 4cdd1b788a Add node conformance ci test. 2016-12-13 19:08:55 -08:00
Random-Liu b7ec229e2c Add run kubelet mode. 2016-12-13 19:08:55 -08:00
Random-Liu bca5aea5ba Refactor RunRemote to support TestSuite interface. 2016-12-13 19:08:55 -08:00
Random-Liu 99dc80ccc2 Add TestSuite interface and update the CreateTestArchive function. 2016-12-13 19:08:55 -08:00
Kubernetes Submit Queue aca523f586 Merge pull request #38664 from dashpole/flaky_inode
Automatic merge from submit-queue

Inode Eviction Test is Flaky

This Pull Request:
Marks the InodeEviciton test as flaky
Increases the timeout for disk pressure because coreos has nearly 2 million inodes.
Decreases the status polling interval so we can see eviction ordering better.
@Random-Liu
2016-12-13 16:22:40 -08:00
David Ashpole 0ec482308e increase timeout and marked as flaky. decreased polling interval to better monitor eviction ordering 2016-12-12 17:31:24 -08:00
Mike Danese c87de85347 autoupdate BUILD files 2016-12-12 13:30:07 -08:00
Random-Liu b27776a25f Update CVM version to e2e-node-containervm-v20161208-image. 2016-12-12 01:50:37 -08:00
Random-Liu 486ddae35a `make test-e2e-node` runs the same test with pr builder by default. 2016-12-09 16:06:18 -08:00
Wojciech Tyczynski a9ec31209e GetOptions - fix tests 2016-12-09 09:42:01 +01:00
Jun Gong 036899ec98 Add --image-pull-progress-deadline option to kubelet 2016-12-09 09:28:57 +08:00
David Ashpole 93ca4bbf47 adjusted timeouts; fixed but where pods are deleted late; added delay to account for delayed pressure 2016-12-08 14:01:30 -08:00
Kubernetes Submit Queue 6b9a944285 Merge pull request #36637 from resouer/nits-e2e
Automatic merge from submit-queue

Fix useless uuid  in container log path node e2e

@timstclair pointed out there're nits in original PR, ref: https://github.com/kubernetes/kubernetes/pull/34877

So this patch: 
1. removed useless uuid
2. change all those strings to const

Thanks. 🐱
2016-12-08 00:43:57 -08:00
Tim St. Clair f4706464f6
Decrease expected lower bound for misc CPU 2016-12-07 12:39:48 -08:00
Random-Liu 6f92572209 Move ssh related functions into ssh.go. 2016-12-05 23:59:58 -08:00
Random-Liu 7c2b1f4752 Remove setup-node, which is not needed after we run the whole test as
root.
2016-12-05 16:16:08 -08:00
Dr. Stefan Schimanski 24e24fc7bb Add verb support to gc and namespace controllers 2016-12-05 12:36:05 +01:00
Clayton Coleman 3454a8d52c
refactor: update bazel, codec, and gofmt 2016-12-03 19:10:53 -05:00
Clayton Coleman 5df8cc39c9
refactor: generated 2016-12-03 19:10:46 -05:00
Kubernetes Submit Queue 9f66f7544b Merge pull request #37856 from timstclair/summary-test
Automatic merge from submit-queue (batch tested with PRs 37692, 37785, 37647, 37941, 37856)

Verify misc container in summary test

Should detect issue from https://github.com/kubernetes/kubernetes/issues/35214, https://github.com/kubernetes/kubernetes/issues/37453

/cc @piosz @dchen1107
2016-12-03 11:45:03 -08:00
Kubernetes Submit Queue ba62dafe39 Merge pull request #37663 from Random-Liu/fix-node-e2e-firewall-configure
Automatic merge from submit-queue (batch tested with PRs 37094, 37663, 37442, 37808, 37826)

Node E2E: Fix node e2e firewall configure.

Get rid of the misleading error message:
```
W1129 12:57:16.967] E1129 12:57:16.967130   29815 remote.go:204] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.201.208 -- sudo sh -c 'iptables -L INPUT | grep "Chain INPUT (policy DROP)"&&(iptables -C INPUT -w -p TCP -j ACCEPT || iptables -A INPUT -w -p TCP -j ACCEPT)&&(iptables -C INPUT -w -p UDP -j ACCEPT || iptables -A INPUT -w -p UDP -j ACCEPT)&&(iptables -C INPUT -w -p ICMP -j ACCEPT || iptables -A INPUT -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
W1129 12:57:17.271] E1129 12:57:17.271169   29815 remote.go:213] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.201.208 -- sudo sh -c 'iptables -L FORWARD | grep "Chain FORWARD (policy DROP)" > /dev/null&&(iptables -C FORWARD -w -p TCP -j ACCEPT || iptables -A FORWARD -w -p TCP -j ACCEPT)&&(iptables -C FORWARD -w -p UDP -j ACCEPT || iptables -A FORWARD -w -p UDP -j ACCEPT)&&(iptables -C FORWARD -w -p ICMP -j ACCEPT || iptables -A FORWARD -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
W1129 12:57:17.557] E1129 12:57:17.556683   29815 remote.go:204] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.128.178 -- sudo sh -c 'iptables -L INPUT | grep "Chain INPUT (policy DROP)"&&(iptables -C INPUT -w -p TCP -j ACCEPT || iptables -A INPUT -w -p TCP -j ACCEPT)&&(iptables -C INPUT -w -p UDP -j ACCEPT || iptables -A INPUT -w -p UDP -j ACCEPT)&&(iptables -C INPUT -w -p ICMP -j ACCEPT || iptables -A INPUT -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
W1129 12:57:17.771] I1129 12:57:17.771236   29815 remote.go:231] Killing any existing node processes on tmp-node-e2e-a1212c32-gci-dev-56-8977-0-0
W1129 12:57:17.877] E1129 12:57:17.877123   29815 remote.go:213] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.128.178 -- sudo sh -c 'iptables -L FORWARD | grep "Chain FORWARD (policy DROP)" > /dev/null&&(iptables -C FORWARD -w -p TCP -j ACCEPT || iptables -A FORWARD -w -p TCP -j ACCEPT)&&(iptables -C FORWARD -w -p UDP -j ACCEPT || iptables -A FORWARD -w -p UDP -j ACCEPT)&&(iptables -C FORWARD -w -p ICMP -j ACCEPT || iptables -A FORWARD -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
W1129 12:57:17.898] I1129 12:57:17.898711   29815 remote.go:239] Extracting tar on tmp-node-e2e-a1212c32-gci-dev-56-8977-0-0
W1129 12:57:17.941] E1129 12:57:17.941566   29815 remote.go:204] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.154.237 -- sudo sh -c 'iptables -L INPUT | grep "Chain INPUT (policy DROP)"&&(iptables -C INPUT -w -p TCP -j ACCEPT || iptables -A INPUT -w -p TCP -j ACCEPT)&&(iptables -C INPUT -w -p UDP -j ACCEPT || iptables -A INPUT -w -p UDP -j ACCEPT)&&(iptables -C INPUT -w -p ICMP -j ACCEPT || iptables -A INPUT -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
W1129 12:57:18.020] I1129 12:57:18.019802   29815 remote.go:231] Killing any existing node processes on tmp-node-e2e-a1212c32-coreos-alpha-1122-0-0-v20160727
W1129 12:57:18.024] E1129 12:57:18.024044   29815 remote.go:213] Failed to configured firewall: command [ssh -i /home/jenkins/.ssh/google_compute_engine -o UserKnownHostsFile=/dev/null -o IdentitiesOnly=yes -o CheckHostIP=no -o StrictHostKeyChecking=no -o ServerAliveInterval=30 -o LogLevel=ERROR 104.154.154.237 -- sudo sh -c 'iptables -L FORWARD | grep "Chain FORWARD (policy DROP)" > /dev/null&&(iptables -C FORWARD -w -p TCP -j ACCEPT || iptables -A FORWARD -w -p TCP -j ACCEPT)&&(iptables -C FORWARD -w -p UDP -j ACCEPT || iptables -A FORWARD -w -p UDP -j ACCEPT)&&(iptables -C FORWARD -w -p ICMP -j ACCEPT || iptables -A FORWARD -w -p ICMP -j ACCEPT)'] failed with error: exit status 1 output: 
```

The problem is that the command 'iptables -L FORWARD | grep "Chain FORWARD (policy DROP)" returns an error when the rule is not found, which is not expected behaviour.

@freehan
2016-12-03 04:27:48 -08:00
Kubernetes Submit Queue 7ec3be4c8e Merge pull request #36964 from ixdy/gobin-build
Automatic merge from submit-queue

Build vendored copy of go-bindata and use that in go generate step

**What this PR does / why we need it**: as the title says, uses the vendored version of `go-bindata` rather than expecting developers to `go get` it (when building outside docker).

**Which issue this PR fixes**: fixes #34067, partially addresses #36655

**Special notes for your reviewer**: we still call `go generate` far too many times:
```console
~/.../src/k8s.io/kubernetes $ which go-bindata
~/.../src/k8s.io/kubernetes $ make
+++ [1116 17:35:28] Building the toolchain targets:
    k8s.io/kubernetes/hack/cmd/teststale
    k8s.io/kubernetes/vendor/github.com/jteeuwen/go-bindata/go-bindata
+++ [1116 17:35:29] Generating bindata:
    test/e2e/framework/gobindata_util.go
+++ [1116 17:35:30] Building go targets for linux/amd64:
    cmd/libs/go2idl/deepcopy-gen
+++ [1116 17:35:35] Building the toolchain targets:
    k8s.io/kubernetes/hack/cmd/teststale
    k8s.io/kubernetes/vendor/github.com/jteeuwen/go-bindata/go-bindata
+++ [1116 17:35:35] Generating bindata:
    test/e2e/framework/gobindata_util.go
+++ [1116 17:35:36] Building go targets for linux/amd64:
    cmd/libs/go2idl/defaulter-gen
+++ [1116 17:35:41] Building the toolchain targets:
    k8s.io/kubernetes/hack/cmd/teststale
    k8s.io/kubernetes/vendor/github.com/jteeuwen/go-bindata/go-bindata
+++ [1116 17:35:41] Generating bindata:
    test/e2e/framework/gobindata_util.go
+++ [1116 17:35:42] Building go targets for linux/amd64:
    cmd/libs/go2idl/conversion-gen
+++ [1116 17:35:47] Building the toolchain targets:
    k8s.io/kubernetes/hack/cmd/teststale
    k8s.io/kubernetes/vendor/github.com/jteeuwen/go-bindata/go-bindata
+++ [1116 17:35:47] Generating bindata:
    test/e2e/framework/gobindata_util.go
+++ [1116 17:35:48] Building go targets for linux/amd64:
    cmd/libs/go2idl/openapi-gen
+++ [1116 17:35:56] Building the toolchain targets:
    k8s.io/kubernetes/hack/cmd/teststale
    k8s.io/kubernetes/vendor/github.com/jteeuwen/go-bindata/go-bindata
+++ [1116 17:35:56] Generating bindata:
    test/e2e/framework/gobindata_util.go
```
Fixing that is a separate effort, though.

cc @sebgoa @ZhangBanger
2016-12-02 07:29:01 -08:00
Kubernetes Submit Queue 3a4b4749e3 Merge pull request #37648 from Random-Liu/collect-serial-output-node-e2e
Automatic merge from submit-queue

Node E2E: Collect serial output

This is a temporary solution to collect serial output from test GCE node in node e2e.

We should come up with a better idea later. Ideally, node e2e should share the same log collection logic with cluster e2e. https://github.com/kubernetes/kubernetes/blob/master/cluster/log-dump.sh

Mark v1.5 because this helps debug https://github.com/kubernetes/kubernetes/issues/37333.

@mtaufen @dchen1107 
/cc @kubernetes/sig-node
2016-12-01 15:06:59 -08:00
Tim St. Clair bcf5e434fa
Verify misc container in summary test 2016-12-01 14:02:39 -08:00