github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
100,590 Commits 1 Branch 31 Tags 1,019 MiB
a8f68f9aba
Commit Graph

19445 Commits

Author SHA1 Message Date
Kubernetes Prow Robot
8d949052c8
Merge pull request #101582 from tsmetana/k8s-fix-aws-kms-key-test 
Fix invalid AWS KMS key test flake
 2021-04-29 18:10:52 -07:00
David Porter
28acfab74c Update cos-gpu-installer image 
This installer image should be compatible with COS-M85 images,
specifically `cos-85-13310-1041-9`
 2021-04-29 14:58:47 -07:00
Kubernetes Prow Robot
204ff6caeb
Merge pull request #101586 from liggitt/delete-partition-tests 
Remove inactive failing tests
 2021-04-29 11:36:23 -07:00
Kubernetes Prow Robot
eb326fdc07
Merge pull request #100262 from chaitanyabandi/daemon-ns-patch 
use Patch API to add namespace annotations
 2021-04-29 11:36:11 -07:00
Kubernetes Prow Robot
19c7089245
Merge pull request #101569 from wzshiming/test/e2e-node-shutdown 
Fix flake for GracefulNodeShutdown e2e
 2021-04-29 09:01:25 -07:00
Michael Taufen
0b3b99c096 Try both in-cluster and external discovery 
The conformance test for ServiceAccountIssuerDiscovery is currently
configured with --in-cluster-discovery, which only supports token
validation against in-cluster endpoints. Many cloud providers provide
their own, external endpoints for OIDC discovery, and because the iss
claim in tokens will point to these endpoints, but the client in this
test only trusts the Cluster CA, it will fail to connect to the external
discovery endpoints when validating the token.

To ensure that the conformance test at least supports scenario where
both the discovery doc endpoint and JWKS endpoint are cluster-local and
the scenario where both endpoints are cluster-external, this PR has the
test try both and requires at least one to pass.

Caveat: The test still won't support a configuration where one
endpoint is cluster-local and the other is external. We don't yet have
evidence that this is a configuration that is used in practice, so this
initial hotfix will at least fix the conformance test for the "both
external" configuration we know providers already use. Note that if one
endpoint is cluster-local, and the other is cluster-external, tokens can
still only be validated in-cluster, because both endpoints must be
accessible to Relying Parties that validate tokens.
 2021-04-29 08:56:02 -07:00
Kubernetes Prow Robot
6e29545168
Merge pull request #99998 from matthyx/deflake-startupprobe 
Deflake startupProbe e2e test
 2021-04-29 01:16:50 -07:00
Kubernetes Prow Robot
db70c2a96a
Merge pull request #99348 from chymy/e2e-rck-unused 
Remove unused code in test/e2e/{upgrades,storage,network}
 2021-04-28 21:02:50 -07:00
Shiming Zhang
91beb10aa4 Fix flake for GracefulNodeShutdown e2e 2021-04-29 10:52:00 +08:00
Kubernetes Prow Robot
e6b4fa3811
Merge pull request #101352 from ahg-g/ahg-nss-integration 
Added integration test for pod affinity namespace selector
 2021-04-28 08:00:50 -07:00
Jordan Liggitt
94a8c7af15 Remove inactive failing tests 2021-04-28 10:03:07 -04:00
Tomas Smetana
8773e0d491 Fix invalid AWS KMS key test flake 2021-04-28 13:22:49 +02:00
Matthias Bertschy
bcd582030f Deflake startupProbe e2e test 
Co-authored-by: Sergey Kanzhelev <S.Kanzhelev@live.com>
 2021-04-28 07:07:01 +02:00
Kubernetes Prow Robot
1cab0adcd5
Merge pull request #101487 from deads2k/annotate-A 
add --all-namespaces to kubectl annotate,label
 2021-04-27 20:56:49 -07:00
Kubernetes Prow Robot
21ab9c4175
Merge pull request #95286 from SergeyKanzhelev/removePodUnknown 
Deprecating PodUnknown podPhase
 2021-04-27 11:21:38 -07:00
Abdullah Gharaibeh
ebf0bca3ce Added integration test for pod affinity namespace selector 2021-04-27 14:13:08 -04:00
Kubernetes Prow Robot
3f98b98c69
Merge pull request #101488 from dcbw/e2e-net-firewall 
e2e/network/firewall: don't assume nodes are exposed externally
 2021-04-26 22:15:36 -07:00
Kubernetes Prow Robot
17f3990ea1
Merge pull request #100484 from gavinfish/e2e-storage-suffix 
Remove suffixes for VolumeSnapshotClasses in E2E tests
 2021-04-26 17:37:03 -07:00
David Eads
ad0cff7b61 add --all-namespaces to kubectl annotate,label 2021-04-26 13:04:19 -04:00
Dan Williams
f86ddbea70 e2e/network/firewall: don't assume nodes are exposed externally 
If no nodes have NodeExternalIP addresses, then clearly none of the
services are exposed externally, and the test should succeed.
Seen in OpenShift CI.
 2021-04-26 10:25:21 -05:00
Kubernetes Prow Robot
2b54418d66
Merge pull request #100387 from jsturtevant/test-image-doc-updates 
Doc updates to clarify the image promotion process
 2021-04-26 06:25:01 -07:00
Kubernetes Prow Robot
3e71ecc6b3
Merge pull request #101329 from ahg-g/ahg-nss-bench 
Added benchmarks for pod affinity NamespaceSelector
 2021-04-25 19:07:00 -07:00
Kubernetes Prow Robot
5c34712a09
Merge pull request #101421 from yuzhiquan/typo 
Fix typo for kubelet
 2021-04-25 12:19:00 -07:00
Artyom Lukianov
79dbdbb4c1 e2e node: fix the race condition under the memory manager test 
Wait for kubelet to be healthy after the dynamic update
of the kubelet configuration.

Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
 2021-04-25 15:12:12 +03:00
Kubernetes Prow Robot
113f828553
Merge pull request #101442 from aramase/ds-cleanup 
update dual-stack test description to remove duplicate tag
 2021-04-24 13:57:00 -07:00
Kubernetes Prow Robot
c1418fff4a
Merge pull request #101402 from wangyx1992/timeSub 
cleanup: replace x.Sub(time.Now()) with time.Until(x) in e2e test
 2021-04-24 05:54:59 -07:00
Kubernetes Prow Robot
10f7e189c2
Merge pull request #100496 from jpbetz/extract-subresources 
Add subresource support to client-go apply extract
 2021-04-23 19:27:43 -07:00
Kubernetes Prow Robot
3029808c91
Merge pull request #99772 from pacoxu/patch-4 
job-e2e: wait exceeds active deadline for 15s
 2021-04-23 19:27:35 -07:00
Anish Ramasekar
24f37695e5
update dual-stack test description to remove duplicate tag 
Signed-off-by: Anish Ramasekar <anish.ramasekar@gmail.com>
 2021-04-23 16:32:36 -07:00
Kubernetes Prow Robot
b11f135a7a
Merge pull request #101381 from prameshj/port-change 
Use a less-common port for e2e test http pod.
 2021-04-23 16:14:29 -07:00
Kubernetes Prow Robot
8846859f77
Merge pull request #99154 from bertinatto/fix-make-pod 
e2e: MakePod() should support block and inline volumes
 2021-04-23 11:21:39 -07:00
Abdullah Gharaibeh
6988653457 Added benchmarks for pod affinity namespaceselector 2021-04-23 14:14:38 -04:00
yuzhiquan
02c3d53a23 typo 2021-04-23 17:55:54 +08:00
wangyx1992
fda7421f24 cleanup: replace x.Sub(time.Now()) with time.Until(x) in e2e test 
Signed-off-by: wangyx1992 <wang.yixiang@zte.com.cn>
 2021-04-23 11:27:12 +08:00
Kubernetes Prow Robot
2365d1e8bd
Merge pull request #101383 from msau42/update-mock 
Update sidecars for csi mock driver
 2021-04-22 17:44:40 -07:00
Kubernetes Prow Robot
66d27ff1a4
Merge pull request #101380 from kwiesmueller/flake-apply-mutating-webhook 
[flake] fix TestMutatingWebhookResetsInvalidManagedFields flakes
 2021-04-22 17:44:31 -07:00
Kubernetes Prow Robot
0e05558717
Merge pull request #101360 from pohly/hostpath-update 
Revert "storage e2e: disable health check containers"
 2021-04-22 17:44:14 -07:00
Kubernetes Prow Robot
2f6d79ca88
Merge pull request #100548 from mauriciopoppe/ephemeral-storage-e2e-windows 
Make ephemeral volume tests work with Windows nodes
 2021-04-22 17:44:01 -07:00
Joe Betz
d01788cd56 Enable extract test for status subresource 2021-04-22 15:32:45 -07:00
Kevin Wiesmueller
96a8df455c fix TestMutatingWebhookResetsInvalidManagedFields flakes 2021-04-23 00:09:31 +02:00
Michelle Au
b67e1b8dfb Update sidecars for csi mock driver 
Change-Id: I751dc41facd95219604124f3f2f221f4d35ca34a
 2021-04-22 14:29:30 -07:00
Kubernetes Prow Robot
0f4de95e5b
Merge pull request #101330 from deads2k/apply-remove-beta 
apply test for manager limit to only use v1 apis
 2021-04-22 13:19:49 -07:00
Kubernetes Prow Robot
3061f6149f
Merge pull request #101171 from KnVerey/kubectl-describe-chunk-size 
Add `--chunk size` flag to `kubectl describe`
 2021-04-22 13:19:23 -07:00
Pavithra Ramesh
2430f23445 Use a less-common port for e2e test http pod. 
8080 is a common port that is more likely to cause conflicts when used
in hostNetwork mode for tests.
 2021-04-22 13:17:10 -07:00
Kubernetes Prow Robot
714a821f1b
Merge pull request #101295 from eddiezane/ez/fix-kubectl-error-code-e2e-test 
Fix kubectl flake: Simple pod should return command exit codes
 2021-04-22 11:27:59 -07:00
Eddie Zaneski
1bc5364b16
[e2e] match kubectl run timeout to pod start timeout 
Signed-off-by: Eddie Zaneski <eddiezane@gmail.com>
 2021-04-22 11:10:56 -06:00
Kubernetes Prow Robot
1b08dde41f
Merge pull request #101191 from tanjing2020/container_manager_test 
Agnhost image's progress name is called agnhost, not test-webserver
 2021-04-22 08:59:53 -07:00
Kubernetes Prow Robot
972ee2d425
Merge pull request #101034 from verb/1.22-ec-api 
Switch alpha Pod ephemeralcontainers API to use Pod kind
 2021-04-22 06:21:41 -07:00
Kubernetes Prow Robot
82d8d08c78
Merge pull request #101280 from huffmanca/address-snapshotting-flakes 
Force NodeUnstageVolume to finish for all distros
 2021-04-22 02:32:31 -07:00
Kubernetes Prow Robot
f1e51ceb3c
Merge pull request #98377 from nodo/scale-ownership-tracking-deployment 
Track ownership of scale subresource
 2021-04-22 01:28:32 -07:00
Patrick Ohly
3299469437 Revert "storage e2e: disable health check containers" 
This reverts commit 0c2cee5676e64976f9e767f40c4c4750a8eeb11f.

The health check containers are not required for any test, but we want
to run them anyway to ensure that they cause no unexpected issues.
 2021-04-22 08:20:39 +02:00
Kubernetes Prow Robot
1f87e3fe00
Merge pull request #101347 from verult/subpath-configmap-timeout 
Extend pod start timeout to 5min for storage subpath configmap test
 2021-04-21 21:12:46 -07:00
Kubernetes Prow Robot
d907c29956
Merge pull request #100637 from pohly/hostpath-update 
storage e2e: automate hostpath YAML updates, update sidecars but not driver
 2021-04-21 21:12:33 -07:00
Cheng Xing
4cfb6e2c23 Extend pod start timeout to 5min for storage subpath configmap test 2021-04-21 17:24:09 -07:00
Mike Dame
07029c941a Remove Limits from scheduling e2e balanced pod resources 
The purpose of the pod created by `createBalancedPodForNodes()` is to ensure
that all nodes have equal resource requests (as seen by the scheduler). This
prevents the default scheduling behavior (which attempts to balance resource requests)
from interfering with e2e's which test other priorities/score plugins.

Because the scheduler only worries about requests, specifying `Limits` in this pod
is unnecessary. In fact, if the calculated "balancing" limit is too low, it can cause
the balancing pod to never start due to OOMKill errors, leading to flakes and failures.
 2021-04-21 15:58:00 -04:00
Kubernetes Prow Robot
1e4388964e
Merge pull request #101293 from ehashman/issue-93480 
Remove unmaintained elasticsearch tests
 2021-04-21 12:20:12 -07:00
Christian Huffman
21ea9af37f Force NodeUnstageVolume to finish for all distros 2021-04-21 15:01:41 -04:00
Katrina Verey
8ad6fd6ddb Add --chunk-size support to kubectl describe 2021-04-21 11:44:15 -07:00
David Eads
b1ce579517 apply test for manager limit to only use v1 apis 2021-04-21 14:38:24 -04:00
Andrea Nodari
8e4b5c849b Do not add managed fields if a scale entry doesn't own replicas 
This happens when a request changes the .status.replicas but not
.spec.replicas
 2021-04-21 20:27:45 +02:00
Andrea Nodari
09649e58b5 Check request info when updating managed fields during scale 
- Test all versions to make sure each resource version is in the
  mappings
- Fail when request info contains an unrecognized version. We have tests
  that guarantee that all known versions are in the mappings. If we
  get a version in request info that is not there we should fail fast to
  prevent inconsistent behaviour (e.g. for some reason the mappings is
  not up to date).

Ensure all known versions are in mappings
 2021-04-21 20:27:02 +02:00
Antoine Pelisse
816e80206c Use ScaleHandler for all scalable resources 2021-04-21 18:41:40 +02:00
Andrea Nodari
a9ea98b3b9 Track ownership of deployments scale subresource 2021-04-21 18:41:40 +02:00
Kubernetes Prow Robot
032007e007
Merge pull request #101312 from harche/ContainerLogPath_fix 
Add SELinux security context to ContainerLogPath test
 2021-04-21 09:31:17 -07:00
Gautier Delorme
34b0fcef5f remove go-openapi/spec 
Signed-off-by: Gautier Delorme <gautier.delorme@gmail.com>
 2021-04-21 11:31:49 +02:00
Fabio Bertinatto
6a645592fc e2e: MakePod() should support block and inline volumes 2021-04-21 11:29:35 +02:00
Harshal Patil
df13eebfd0 Add SELinux security context to ContainerLogPath test 
Signed-off-by: Harshal Patil <harpatil@redhat.com>
 2021-04-21 13:48:32 +05:30
Patrick Ohly
c794b5c442 storage e2e: patch in RBAC rules for secrets 
In one mock test, the snapshotter needs permission to read
secrets. That was disabled in the RBAC files of recent releases. We
need to patch it back in during deployment.
 2021-04-21 09:57:54 +02:00
Shiming Zhang
e08988ba16 Fix pull empty image URL 
Signed-off-by: Shiming Zhang <wzshiming@foxmail.com>
 2021-04-21 15:29:46 +08:00
Kubernetes Prow Robot
497e7c0c1e
Merge pull request #101233 from pohly/e2e-suite-cleanup 
storage e2e: remove obsolete function
 2021-04-20 19:24:10 -07:00
Kubernetes Prow Robot
6a7572e4ad
Merge pull request #101287 from soltysh/replace_with_default 
Fail fast when image is specified without tag
 2021-04-20 17:38:24 -07:00
Elana Hashman
ac5ba850c1
Remove unused elasticsearch tests 2021-04-20 13:34:59 -07:00
Maciej Szulik
3edabfb460
Switch to a proper busybox test image 2021-04-20 22:05:31 +02:00
Maciej Szulik
89b50bcf94
Fail fast when image is specified w/o tag 2021-04-20 21:10:16 +02:00
Kubernetes Prow Robot
92aff21558
Merge pull request #95609 from fromanirh/tm-e2e-faster-delete 
e2e: topology manager: use deletePodSync for faster delete
 2021-04-20 04:14:33 -07:00
Patrick Ohly
ebd02341c9 storage e2e: downgrade hostpath driver 
This is a temporary workaround until a fixed driver is available.
 2021-04-20 08:07:15 +02:00
Patrick Ohly
7682e39a47 storage e2e: disable health check containers 
They are not needed for any of the tests and in practice apparently
caused enough overhead that even unrelated tests timed out. For
example, in the pull-kubernetes-e2e-kind test, 43 out of 5771 tests
failed, including tests from sig-node, sig-cli, sig-api-machinery,
sig-network.
 2021-04-20 08:07:15 +02:00
Patrick Ohly
446c1136dc storage e2e: automate hostpath YAML updates, hostpath v1.6.2 
Mirroring the various YAML files by hand is tedious. The new
update-hostpath.sh does all the necessary steps automatically.

The result is now a bit more consistent with the upstream repos in the
sense that the original file names and paths for the RBAC YAML files
are used.

The csi-hostpath-testing.yaml is included for the sake of
completeness, but not used during E2E testing.

The new hostpath driver release is v1.6.2, which adds the
external-health-monitor for the first time.
 2021-04-20 08:07:15 +02:00
Kubernetes Prow Robot
0f5fce6a64
Merge pull request #101244 from deads2k/remove-csr 
remove CSR v1beta1 tests since the API is no longer available in 1.22 and CSRs are short term
 2021-04-19 22:12:39 -07:00
Kubernetes Prow Robot
1eccb41fa8
Merge pull request #101248 from deads2k/integration 
update more integration tests for v1beta1 removal
 2021-04-19 18:54:56 -07:00
Kubernetes Prow Robot
651a7a6f6b
Merge pull request #101245 from deads2k/test-cmd 
update test-cmd to avoid removed beta APIs
 2021-04-19 18:54:44 -07:00
James Sturtevant
02a182f5e2 updates to clarify the image promotion process 2021-04-19 15:49:27 -07:00
David Eads
e6113c5fe6 update openapi overlap test for crd v1 2021-04-19 13:55:32 -04:00
David Eads
b199047fa6 remove outdated extensions group test 2021-04-19 13:55:32 -04:00
David Eads
b74c542de5 skip printing test for removed APIs 2021-04-19 13:55:32 -04:00
Shihang Zhang
925900317e allow multiple of --service-account-issuer 2021-04-19 09:54:11 -07:00
David Eads
2fb8420756 update test-cmd to avoid removed beta APIs 2021-04-19 10:06:37 -04:00
David Eads
2e07319356 remove CSR v1beta1 tests since the API is no longer avaiable and CSRs are short term 2021-04-19 09:39:41 -04:00
Patrick Ohly
93d1e373eb storage e2e: remove obsolete function 
SkipRedundantSuite was replaced by SkipUnsupportedTests. We don't need
both.
 2021-04-19 11:11:47 +02:00
Kubernetes Prow Robot
9f5573e83f
Merge pull request #101211 from aojea/nosnat 
e2e create pods only on schedulable nodes
 2021-04-17 10:32:34 -07:00
Kubernetes Prow Robot
0f1d105f8d
Merge pull request #100970 from apelisse/add-subresource-managedfields 
Add subresource managedfields
 2021-04-17 07:42:34 -07:00
Antonio Ojea
0695bc45cf e2e create pods only on schedulable nodes 
We were avoiding the scheduled using the pod.Spec.NodeName directly,
however, once we switched to using the node selector, the no_snat
e2e test started to fail because was trying to schedule pods on
nodes with taints, hence, failing the test.
 2021-04-17 12:05:40 +02:00
Stephen Augustus
fac97f8a9b [go1.16] Update to go1.16.3 
Signed-off-by: Stephen Augustus <foo@auggie.dev>
 2021-04-16 19:56:51 -04:00
Antoine Pelisse
e477c31307 fieldmanager: Comment test broken by status being separated from main resource 2021-04-16 14:26:05 -07:00
Andrea Nodari
862d256195 Add "subresource" field to ManagedFieldEntry 
This field is useful to namespace the managed field entries of a
subresource and differentiate them from the ones of the main resource.
 2021-04-16 14:13:18 -07:00
Kubernetes Prow Robot
3ed71cf190
Merge pull request #100976 from jindijamie/master 
releng: Update debian-base and debian-iptables to buster-v1.6.0 to patch base image CVEs
 2021-04-16 12:56:34 -07:00
Kubernetes Prow Robot
f1c1379def
Merge pull request #100963 from enj/enj/i/authz_func_ctx 
authorizer func: pass through context
 2021-04-16 12:56:22 -07:00
Kubernetes Prow Robot
27a625cf89
Merge pull request #100724 from liggitt/eviction-v1beta1 
Add policy/v1 Eviction support
 2021-04-16 10:02:22 -07:00
Antonio Ojea
60c3480740 e2e don't set pod.Spec.NodeName directly 
based on this comment in
ea07644522/test/e2e/framework/pod/node_selection.go (L96-L101)

// pod.Spec.NodeName should not be set directly because
// it will bypass the scheduler, potentially causing
// kubelet to Fail the pod immediately if it's out of
// resources. Instead, we want the pod to remain
// pending in the scheduler until the node has resources
// freed up.
 2021-04-16 15:58:16 +02:00
Lee Verberne
97726a50c1 Address feedback for new /ephemeralcontainers API 
* Use deep copies in `PrepareForUpdate()`
* Preserve select metadata from new pod
* Use patch to add ephemeral container `kubectl debug`
* Distinguish between pod vs /ephemeralcontainers NotFound
 2021-04-16 14:08:12 +02:00