Commit Graph

3822 Commits

Author SHA1 Message Date
k8s-merge-robot
22d405477b Merge pull request #26568 from derekwaynecarr/kubectl_qos
Automatic merge from submit-queue

Update kubectl describe pod to use pod level qos

Fixes https://github.com/kubernetes/kubernetes/issues/26374

/cc @vishh
2016-06-04 02:40:47 -07:00
k8s-merge-robot
707cc2bbb8 Merge pull request #26493 from caesarxuchao/fix-gc-flake
Automatic merge from submit-queue

Fixes 25890 flake. Let GC convert ListOptions to v1 before passing it to the dynamic client

GC's ListWatcher directly passed the api.ListOptions to the dynamic client, but the parameter codec of dynamic client converts the options to queries based on the tags in the struct, which are not present in api.ListOptions, so the queries are not sent to the server. As a result, the Watch request was sent without a resourceVersion, causing missed events. Flake #25890 is caused by the missed deletion events.

This PR converts the api.ListOptions to v1.ListOptions before the GC passes it to the dynamic codec. The flaky test has successfully passed 79 times ([log](https://00e9e64bacd064560a027fbee9c5a373a1614f3a56e652ae40-apidata.googleusercontent.com/download/storage/v1_internal/b/kubernetes-jenkins/o/pr-logs%2Fpull%2F25923%2Fkubernetes-pull-test-unit-integration%2F28364%2Fbuild-log.txt?qk=AD5uMEv72OjSUqDyk5i-ZLurcmM4i7gket1c7WaqR7yuIYz7WhPYT7ewVBafijV0ymnPTYqxRYt1kp6S9YQv7chPwC-3UtrKetKfhYnvAFrPGXAIBxHytTmpFohRAYgsARN1B6j1f9vyK5lM-8jyzRGhCK3sCRsAPnbDBWIWFlbH4b1n3vUET3P71QamHrF5itYyaqRU5pMZV3Cwwr81X8q7h5hCzm3Ip78RpMzfjEqTG0RcM2TLGccUrlkWVBLh4hn0NFpUIkzVFugFA5ooJffo-0AdJnO3mGWEOnXNVFWftJbK8cKnTns0DISrYFOyH_PlOe_YHCxgIXIT-dW8G-nbqoUjn5SBqunr36rcpaYCIwe2va4W_AcLCT43xiEAezRER_U9AuIqi_22KMd6SuHTyljhmWFPvPk8-gpjthLWXhcE7LPO5dV41hnZHnbI4n_9eI1nSVm7q9XdSvX1sWKV1GCwn8oj017AnxVvl9bScultko_0dTC747UqJ6UTFakLuFcHFe-F5Tz7ItDWlBVPoXeC7gTpyuicFKLsdqGlW9F5X6kIwNrBRj9uRsS-QuzSER-fVkQCn4dUTcokttRH_0bYvyfr9oqiDXmywMgOp-L0sKayk8JOVynh2q0Tju9sdkvFr0PxoAjhofomfIC1SZ_JkOzwAT1TUW8dLjPHluMct34xW_-qna1AmkoxM4bZQLhllap96NTC-0IdtzeKDrTul8p7u3WXSJjjEMSijibTNMlnkB0AluT1_RNO94OnzuFv4YlcV24FPhJzchhbyKREkOb_wzgcnSbRwGHjIcfRgkX-IzoXHVBcMYFUrPmsXrnRcfad4XwjkUOgvivkURW2_EwnzgrLDh-IKek51_0FpT1MnFCSG0gQbVSs_iMVPr6UXNAw62LGbKVtl3ZMXyapEpcO8azNbn6Wvd550R704JXxYlU)).

@lavalamp @krousey @smarterclayton
2016-06-04 01:52:31 -07:00
k8s-merge-robot
0418a2c0ad Merge pull request #26153 from jsafrane/intergation-binder-stress
Automatic merge from submit-queue

volume controller: add configurable integration test to stress the binder

The test tries to bind configured nr. of PVs to the same nr. of PVCs. '100' is used by default, which should take ~1-3 seconds (depends on log level). Periodic sync is needed in rare cases, which may add another 10 seconds. - cache from #25881 will help here and sync should not be needed at all.

The test is configurable and may be reused to measure binder performance. Set KUBE_INTEGRATION_PERSISTENTVOLUME_* env. variables as described in persistent_volume_test.go and run the tests:

```
# compile
$ cd test/integration
$ godep go test -tags 'integration no-docker' -c

# run the tests
$ KUBE_INTEGRATION_PERSISTENTVOLUME_SYNC_PERIOD=10s KUBE_INTEGRATION_PERSISTENTVOLUME_OBJECTS=1000 time ./integration.test -test.run TestPersistentVolumeMultiPVsPVCs -v 2
```

Log level '2' is useful to get timestamps of various events like 'TestPersistentVolumeMultiPVsPVCs: start' and 'TestPersistentVolumeMultiPVsPVCs: claims are bound'.
2016-06-04 00:13:29 -07:00
k8s-merge-robot
7d852febc9 Merge pull request #26794 from yujuhong/wait_before_test
Automatic merge from submit-queue

kubelet e2e: enforce that image prepulling must finish before the test

The image prepulling pod calls docker directly to pull images. If the pod
hasn't finished before running the resource usage tracking test, there'd be a
cpu spike in docker. We'd rather wait and fail if this is the case, before
running the test.
2016-06-03 19:42:42 -07:00
k8s-merge-robot
3f1960b73e Merge pull request #26314 from vishh/all-logs
Automatic merge from submit-queue

Link kernel log files to have them be copied over as part of node e2e.

Fixes #25993

cc @pwittrock
2016-06-03 18:49:25 -07:00
Yu-Ju Hong
a6a3ed210c Port the downward api test to the node e2e suite
Also extend the framework to allow a custom client config loading function, so
that the node e2e suite can reuse the same framework across tests.
2016-06-03 18:16:24 -07:00
pwittrock
c74997508e Node e2e use vendored testing packages. 2016-06-03 17:01:50 -07:00
Tim St. Clair
7153981bfe Fix node e2e image pulling retry logic 2016-06-03 16:37:58 -07:00
Vishnu kannan
82718c186c update coreos node e2e images to include updated systemd config
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-03 16:14:26 -07:00
Vishnu kannan
7665120818 update attributes on node e2e copy image script
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-03 16:14:26 -07:00
Vish Kannan
4e1596e61e Remove debug log line 2016-06-03 14:41:05 -07:00
Vishnu kannan
4e3bce8cb7 Mark runtime conformance tests as flaky and not run them in jenkins CI.
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-03 13:57:44 -07:00
k8s-merge-robot
69e45dc30d Merge pull request #26671 from vishh/node-e2e-images
Automatic merge from submit-queue

Support images and instances to exist in different GCP projects for node

Forked off from #26314
2016-06-03 13:38:02 -07:00
derekwaynecarr
89ff3b2530 Update kubectl describe pod to use pod level qos 2016-06-03 15:32:32 -04:00
Yu-Ju Hong
02d7af28e0 kubelet e2e: enforce that image prepulling must finish before the test
The image prepulling pod calls docker directly to pull images. If the pod
hasn't finished before running the resource usage tracking test, there'd be a
cpu spike in docker. We'd rather wait and fail if this is the case, before
running the test.
2016-06-03 12:25:25 -07:00
Chao Xu
06f49f7ca7 Let the dynamic client take a customized parameter codec for List, Watch, and DeleteCollection.
Let the gc's ListWatcher use api.ParameterCodec. Fixes 25890.
2016-06-03 11:22:51 -07:00
Vishnu kannan
34a828d07e Link kernel log files to have them be copied over as part of node e2e.
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-03 11:22:25 -07:00
Jerzy Szczepkowski
a9dc78c6f6 Implemented e2e test: cluster autoscaler with node selector.
Implemented e2e test: cluster autoscaler with node selector.
2016-06-03 16:38:45 +02:00
Jerzy Szczepkowski
07d9e2dcb0 Implemented support for multiple MIGs in e2e cluster autoscaling tests.
Improved check in shouldn't increase cluster size e2e test.
2016-06-03 16:31:56 +02:00
k8s-merge-robot
81c5cfc139 Merge pull request #26773 from mwielgus/gke-autoscaling
Automatic merge from submit-queue

Setup for cluster autoscaling tests in GKE

cc: @piosz @jszczepkowski @fgrzadkowski
2016-06-03 07:31:14 -07:00
Marcin Wielgus
ab56306bec Setup for cluster autoscaling tests in GKE 2016-06-03 14:45:08 +02:00
Jan Safranek
1cfc984b70 volume controller: add configurable integration test to stress the binder
The test tries to bind configured nr. of PVs to the same nr. of PVCs.
'100' is used by default, which should take ~1-3 seconds (depends on log level).
Periodic sync is needed in rare cases, which may add another 10 seconds. - cache
from #25881 will help here and sync should not be needed at all.

The test is configurable and may be reused to measure binder performance.
Set KUBE_INTEGRATION_PV_* env. variables as described in
persistent_volume_test.go and run the tests:

# compile
$ cd test/integration
$ godep go test -tags 'integration no-docker' -c

# run the tests
$ KUBE_INTEGRATION_PV_SYNC_PERIOD=10s KUBE_INTEGRATION_PV_OBJECTS=1000 time ./integration.test -test.run TestPersistentVolumeMultiPVsPVCs -v 2

Log level '2' is useful to get timestamps of various events like
'TestPersistentVolumeMultiPVsPVCs: start' and 'TestPersistentVolumeMultiPVsPVCs:
claims are bound'.
2016-06-03 12:52:32 +02:00
k8s-merge-robot
45aa24b6b7 Merge pull request #26629 from jsafrane/stabilize
Automatic merge from submit-queue

Stabilize persistent volume integration tests

- add more logs
- wait both for volume and claim to get bound

When binding volumes to claims the controller saves PV first and PVC right
after that. In theory, this saved PV could cause waitForPersistentVolumePhase
to finish and PVC could be checked in the test before the controller saves it.
So, wait for both PVC and PV to get bound and check the results only after
that. This is only a theory, there are no usable logs in integration tests.

Fixes #26499 (at least I hope so...)
2016-06-03 03:10:00 -07:00
k8s-merge-robot
4adc00eceb Merge pull request #26454 from thockin/licenses_again
Automatic merge from submit-queue

Don't allow deps with no discernible license

This updates the few deps we had with no LICENSE file to current versions that do have that file.  It also disallows new deps without obvious licenses.
2016-06-03 01:18:14 -07:00
k8s-merge-robot
693dae70a8 Merge pull request #26226 from ArtfulCoder/reversedns
Automatic merge from submit-queue

Added DNS Reverse Record logic for service IPs
2016-06-02 18:43:39 -07:00
Tim Hockin
3c6ce53b7f Fix call-sites after dep updates 2016-06-02 16:57:28 -07:00
k8s-merge-robot
c114f32998 Merge pull request #26289 from derekwaynecarr/enable_accounting
Automatic merge from submit-queue

Enable node e2e accounting on systemd

Updated the e2e setup.sh script to enable cpu and memory accounting.

Related to https://github.com/kubernetes/kubernetes/issues/26198

/cc @pwittrock
2016-06-02 13:42:58 -07:00
k8s-merge-robot
b6a7be6f66 Merge pull request #26533 from gmarek/retries
Automatic merge from submit-queue

Add retries for make in start-kubemark

Fix. #25872
2016-06-02 10:49:28 -07:00
k8s-merge-robot
914163247d Merge pull request #26711 from yujuhong/fix_restart
Automatic merge from submit-queue

e2e: fix node restart tests

The test does not run in the per-PR suite, so e2e is not required.
2016-06-02 08:54:40 -07:00
k8s-merge-robot
cae939b81c Merge pull request #26654 from janetkuo/e2e-deployment-list-rs
Automatic merge from submit-queue

List RSes once when getting old/new RSes in deployment e2e tests

Ref #26509 

[![Analytics](https://kubernetes-site.appspot.com/UA-36037335-10/GitHub/.github/PULL_REQUEST_TEMPLATE.md?pixel)]()
2016-06-02 08:54:36 -07:00
Yu-Ju Hong
8b082ce57d e2e: fix node restart tests
Allow pods to be succeeded or running.
2016-06-02 07:38:51 -07:00
k8s-merge-robot
0aaf5316be Merge pull request #26674 from vishh/node-e2e-owners
Automatic merge from submit-queue

adding a few owners for node e2e framework

cc @pwittrock 

This should help spread node e2e PRs.
2016-06-02 06:52:10 -07:00
Wojciech Tyczynski
a44d027e2e Merge pull request #26695 from gmarek/maxpods
Fix MaxPods test to work with image puller
2016-06-02 14:24:05 +02:00
k8s-merge-robot
0ba4d88e4c Merge pull request #26697 from gmarek/affinity
Automatic merge from submit-queue

Disable PodAffinity SchedulerPredicates test

This feature is disabled, so it's not surprising that tests don't work.

cc @davidopp @kevin-wangzefeng 

@david-mcmahon - this disables the second test that causes failures in SchedulerPredicates suite. When this and #26695 are merged it should be passing in serial.
2016-06-02 05:15:58 -07:00
k8s-merge-robot
61fb527801 Merge pull request #26631 from gmarek/constraints
Automatic merge from submit-queue

Revert revert of adding resource constraints for master components in density tests

The problem was the time when resource constraints were generated. It turns out that the provider is not set there. This version should work.

cc @roberthbailey @alex-mohr
2016-06-02 05:15:55 -07:00
k8s-merge-robot
36e4f3dbe3 Merge pull request #26692 from wojtek-t/fix_scalability_test
Automatic merge from submit-queue

Wait for all nodes to be schedulable in scalability tests
2016-06-02 04:25:04 -07:00
gmarek
262239d7ec Fix MaxPods test to work with image puller 2016-06-02 13:17:45 +02:00
gmarek
72dcf9c236 Disable PodAffinity SchedulerPredicates test 2016-06-02 13:12:39 +02:00
Wojciech Tyczynski
2a5883fc3a Wait for all nodes to be schedulable in scalability tests 2016-06-02 10:20:59 +02:00
Jerzy Szczepkowski
862e2fe2fb Fixed problem with too long name in e2e autoscaling tests.
Fixed problem with too long name in e2e autoscaling tests.
2016-06-02 09:40:11 +02:00
k8s-merge-robot
cf234ab67d Merge pull request #26251 from caesarxuchao/add-serializer
Automatic merge from submit-queue

Add direct serializer

Fix #25589. Implemented a direct codec that doesn't do conversion, but sets the group, version and kind before serialization as Clayton suggested [here](https://github.com/kubernetes/kubernetes/issues/25589#issuecomment-219168009).

First commit is cherry-picked from #24826.

@kubernetes/sig-api-machinery
2016-06-01 23:31:47 -07:00
k8s-merge-robot
8dcedbab5d Merge pull request #26232 from vishh/25713
Automatic merge from submit-queue

Suppress non error logs from ssh commands in node e2e framework.

Fixes #25713
2016-06-01 21:27:29 -07:00
k8s-merge-robot
cdbda4a00f Merge pull request #26645 from yujuhong/cpu_limit
Automatic merge from submit-queue

kubelet e2e: bumping cpu limit

The previous limit was too aggressive and caused kubernetes-e2e-gce-serial build 1404 to fail.
2016-06-01 18:51:03 -07:00
Vishnu kannan
76dc29ef58 adding a few owners for node e2e framework
Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-01 18:11:18 -07:00
Vishnu kannan
60ed27783e Support images and instances to exist in different GCP projects for node
e2e.
Add current user to docker group on instances for node e2e.

Signed-off-by: Vishnu kannan <vishnuk@google.com>
2016-06-01 18:01:02 -07:00
Janet Kuo
2ced966cd5 List RSes once when getting old/new RSes in deployment e2e tests 2016-06-01 14:19:35 -07:00
Alex Robinson
903821354e Merge pull request #26642 from wojtek-t/fix_kubelet_test_panics
Fix panic in kubelet test
2016-06-01 15:41:35 -04:00
k8s-merge-robot
46323336cb Merge pull request #26559 from wojtek-t/max_idle_conns_per_host
Automatic merge from submit-queue

Max idle conns per host

Ref #26519

Hopefully, this will at least somehow help with #26519 

@gmarek @krousey @lavalamp
2016-06-01 11:41:42 -07:00
Yu-Ju Hong
ede3e885d6 kubelet e2e: bumping cpu limit 2016-06-01 11:40:48 -07:00
Wojciech Tyczynski
13b57f8ffb Fix panic in kubelet test 2016-06-01 19:55:35 +02:00