github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
111,859 Commits 1 Branch 31 Tags
0f66366afff2854cfbf60ce422ec0ff69266b995
Commit Graph

22263 Commits

Author SHA1 Message Date
Abu Kashem
aa04af3f5c apf: don't omit flowschemas from reset fields test 2022-09-19 13:46:38 -04:00
Patrick Ohly
004c1b6441 e2e framework: abort polling for running pod early 
When the pod state is failed or completed, the pod will never again enter the
running state and polling can stop.
 2022-09-19 18:10:04 +02:00
Jordan Liggitt
9cfc128b8e Simplify P&F concurrency test 2022-09-19 08:26:29 -04:00
Michal Wozniak
e11039b6f5 Cleanup the TestEvictionForNoExecuteTaintAddedByUser test 2022-09-19 10:07:54 +02:00
Michal Wozniak
ca90621b2a Simplify the integration test for node lifecycle manager 2022-09-19 10:07:35 +02:00
Davanum Srinivas
a1635fa702 fix how we look for our request 
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
 2022-09-18 17:21:19 -04:00
Davanum Srinivas
3eaca7cf51 update code to use newer otel api 
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
 2022-09-17 21:48:20 -04:00
Kubernetes Prow Robot
d2e3c70f1d Merge pull request #111961 from pohly/e2e-klog-output 
e2e: klog output
 2022-09-17 14:38:28 -07:00
Manjunath A Kumatagi
2384a341e1 Add no-op package file to avoid build constraints exclude all Go files error 2022-09-17 12:15:49 +05:30
Kubernetes Prow Robot
3e5e5cc7ee Merge pull request #112357 from SataQiu/deprecated-20220907 
kube-scheduler: add taints filtering logic consistent with TaintToleration plugin for PodTopologySpread plugin
 2022-09-16 18:14:27 -07:00
Kubernetes Prow Robot
4bc1398c08 Merge pull request #112510 from humblec/new-2 
various typo corrections in test/e2e package
 2022-09-16 14:59:00 -07:00
Kubernetes Prow Robot
ea3637f447 Merge pull request #112511 from aojea/e2e_dnspolicy 
add e2e test for hostNetwork pods with dnsPolicy: ClusterFirstWithHostNet
 2022-09-16 12:50:50 -07:00
Kubernetes Prow Robot
06fd0a0728 Merge pull request #112489 from dims/etcd_update_v3.5.5 
etcd: Updated to v3.5.5
 2022-09-16 08:49:16 -07:00
Humble Chirammal
40621d4351 Fix the test failure on volume snapshot 
Since we have upgraded to snapshot controller version to v6, the
snapshot tests looks to be failing in the testgrid. It has been mainly
because the latest version of snapshot controller stopped serving
v1beta1 APIs. The sidecar image versions in the tests also has to be
updated to make sure these are compatible.

This commit add missing RBAC rules for the controller as per the
latest version.

Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
 2022-09-16 19:39:47 +05:30
Kubernetes Prow Robot
740011927c Merge pull request #112504 from pohly/e2e-wait-service-account 
e2e framework: better error when service account creation times out
 2022-09-16 06:31:28 -07:00
Humble Chirammal
9e9fc2be88 various corrections in test/e2e package 
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
 2022-09-16 18:59:30 +05:30
Antonio Ojea
210c1302ef add e2e test for hostNetworl pods with dnsPolicy: ClusterFirstWithHostNet 2022-09-16 15:26:38 +02:00
Kubernetes Prow Robot
0f582f7c3f Merge pull request #112416 from aojea/race_apiserver_startup 
Don't run TestCleanup that requires etcd on non linux+amd64 platforms
 2022-09-16 02:49:18 -07:00
Patrick Ohly
efc172f4bf e2e framework: better error when service account creation times out 
When waiting for the default service account in a new namespace, not finding
one was reported as "unexpected error: timed out waiting for the condition"...
 2022-09-16 11:21:31 +02:00
Antonio Ojea
920dec740b skip etcd unit tests on non fully supported architectures 
etcd only fully supports linux && amd64, the other architectures
and OS are only guaranteed to build, see:

https://etcd.io/docs/v3.5/op-guide/supported-platform/#support-tiers

Skip the test that use etcd on not well supported environment to
guarantee the stability of the test.
 2022-09-16 10:38:03 +02:00
Patrick Ohly
0bd901afea e2e framework: better error messages when pod has failed or completed 
Getting a message about "pod ran to completion" is confusing when the pod
hasn't been able to start at all. The failed state now has a different message.

To address the previous ambiguity, the success state is described as "ran to
completion successfully".
 2022-09-16 09:12:51 +02:00
Davanum Srinivas
0f6b9b883c etcd: Updated to v3.5.5 
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
 2022-09-15 21:05:13 -04:00
Kubernetes Prow Robot
efaf56d4b6 Merge pull request #112444 from piotrnosek/hpamultimplemetrics 
HPA E2E tests: Multiple Metrics
 2022-09-15 16:41:18 -07:00
Kubernetes Prow Robot
9f62f4f087 Merge pull request #112392 from alexkats/enable-tests-for-large-clusters 
Remove 'DisabledForLargeClusters' tag for LB tests
 2022-09-14 13:07:34 -07:00
Kubernetes Prow Robot
4e8b11d441 Merge pull request #111333 from flant/selfsubjectattributesreviews 
Add auth API to get self subject attributes
 2022-09-14 13:07:22 -07:00
Kubernetes Prow Robot
148476eff1 Merge pull request #112427 from gnufied/add-zone-field-cloud-config-test 
Add zone field to vsphere test cloudconfig
 2022-09-14 11:09:25 -07:00
Kubernetes Prow Robot
ee94dce5b1 Merge pull request #112015 from humblec/glusterfs-gone 
Removal of GlusterFS code from the repo
 2022-09-14 10:03:00 -07:00
m.nabokikh
00dfba473b Add auth API to get self subject attributes 
Signed-off-by: m.nabokikh <maksim.nabokikh@flant.com>
 2022-09-14 18:00:26 +02:00
Piotr Nosek
72faedc291 HPA E2E tests: multiple metrics 2022-09-14 11:16:25 +00:00
Kubernetes Prow Robot
9e65786254 Merge pull request #112421 from pohly/e2e-annotated-location 
e2e: improve description of framework callbacks
 2022-09-14 01:09:00 -07:00
Hemant Kumar
30ac00eeb5 Add zone field to vsphere test cloudconfig 2022-09-13 17:19:33 -04:00
Riaan Kleinhans
7ceeca13a3 Fix 1.17 to 1.7 2022-09-14 09:14:28 +12:00
Humble Chirammal
c2a7bfd6b4 removal of glusterfs in-tree driver code from the source 
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
 2022-09-13 20:11:50 +05:30
Patrick Ohly
84990d53cf e2e: improve description of framework callbacks 
When Ginkgo shows a BeforeEach/AfterEach/DeferCleanup, then it can only show
the source code where the callback was registered because there is no
description parameter. This can be improved by passing a custom CodeLocation.

Because a description like "set up framework" might not be enough, the source
code is still shown, too.
 2022-09-13 14:27:32 +02:00
Riaan Kleinhans
f5502818ad Update ineligible_endpoints.yaml 2022-09-13 11:16:42 +12:00
Riaan Kleinhans
4b4084915a Update_ineligible_endpoints.yaml 
remove createCoreV1NamespacedBinding
 2022-09-13 09:35:58 +12:00
Kubernetes Prow Robot
46e1718462 Merge pull request #112351 from piotrnosek/fixbehaviortests 
Fix flaky and failing HPA E2E Behavior tests
 2022-09-12 10:55:37 -07:00
Kubernetes Prow Robot
d7d82ad972 Merge pull request #112252 from piotrnosek/hpa-average-metrics 
Add e2e HPA Tests: CPU and Memory Average with different aggregation methods: Average Value and Utilization
 2022-09-12 09:41:24 -07:00
Kubernetes Prow Robot
eedf0ed221 Merge pull request #112390 from pohly/e2e-storage-helpers 
e2e storage: update helpers
 2022-09-12 08:07:48 -07:00
Kubernetes Prow Robot
48256c55c2 Merge pull request #112251 from pohly/e2e-storage-proxy-shutdown 
e2e storage: close all pending streams when proxy listener gets closed
 2022-09-12 08:07:24 -07:00
Alex Katsman
945aea4627 Remove 'DisabledForLargeClusters' tag for LB tests 
We performed several internal tests at Google both on GKE and
Kubernetes, and we believe it's safe to re-enable them for large
clusters.
 2022-09-12 15:08:42 +02:00
Patrick Ohly
24230b4307 e2e storage: support creating ReplicaSet 
This is useful for running a driver on a subset of all ready nodes:

- use e2enode.GetBoundedReadySchedulableNodes with a suitable
  maximum number of nodes to determine how much nodes are available
  for a test
- define pod anti-affinity in the PodTemplate:
      affinity:
        podAntiAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
          - labelSelector:
              matchLabels:
                app.kubernetes.io/instance: xxxxxxx
            topologyKey: kubernetes.io/hostname
- set the ReplicaSetSpec.Replicas value to the number of nodes
 2022-09-12 14:38:28 +02:00
Patrick Ohly
b6b25c80de e2e storage: add CreateFile to pod IO operations 
This complements the other operations and will be used by some
future tests.
 2022-09-12 14:38:27 +02:00
Kubernetes Prow Robot
3e9e41f82c Merge pull request #112389 from pohly/test-cleanup-flake 
e2e: avoid random control plane output through GinkoWriter
 2022-09-12 05:27:37 -07:00
Patrick Ohly
2e61dfbf1d e2e: avoid random control plane output through GinkoWriter 
If the control plane emits anything at the time when the test runs, for example
"unable to sync kubernetes service", the test breaks because that additional
output is unexpected.
 2022-09-12 12:02:37 +02:00
Patrick Ohly
d33e66b81b e2e: reduce built time for framework 
Pulling the CreateKubeConfig function from the expensive to build
test/utils/apiserver package had a considerable impact on the overall build
time because that package depends on a lot of other packages.

Because only that one function is needed by the framework, that extra build
time can be avoided by moving it into its own package.
 2022-09-10 17:56:23 +02:00
SataQiu
bc9c494319 kube-scheduler: add taints filtering logic consistent with TaintToleration plugin for PodTopologySpread plugin 
Signed-off-by: SataQiu <shidaqiu2018@gmail.com>
 2022-09-10 09:04:30 +08:00
Kubernetes Prow Robot
d9f21e55df Merge pull request #111998 from pohly/e2e-framework-after-each 
e2e framework: replace custom cleanup support with Ginkgo v2 mechanisms
 2022-09-09 16:17:22 -07:00
Piotr Nosek
3d7855392f HPA E2E Test: CPU / Memory: Average Value / Utilization 2022-09-09 16:06:55 +00:00
Kante Yin
096dafe757 Feat: unify the status handle when return in bindingCycle (#112103) 
Signed-off-by: kerthcet <kerthcet@gmail.com>

Signed-off-by: kerthcet <kerthcet@gmail.com>
 2022-09-09 08:31:23 -07:00