Commit Graph

1761 Commits

Author SHA1 Message Date
Artyom Lukianov
a4b367a6a3 Refactor and add new tests to hugepages e2e tests
Add tests to cover usage of multiple hugepages with different
page sizes under the same pod.

Signed-off-by: Artyom Lukianov <alukiano@redhat.com>
2020-06-08 11:23:54 +03:00
Kubernetes Prow Robot
4c8e5c5a50 Merge pull request #91543 from bsdnet/runner
Simplify the logic by removing dead code and enhance logging
2020-06-06 17:29:45 -07:00
Aaron Crickenberger
019d3ee438 update e2e_node OWNERS file
specifically:
- move inactive approvers to emeritus
- add newly active contributors as reviewers
- add a sig/node label to PRs that touch this dir
2020-06-05 10:44:29 -07:00
Kubernetes Prow Robot
8ce1b535ee Merge pull request #80831 from odinuge/hugetlb-pagesizes-cleanup
Add support for removing unsupported huge page sizes
2020-06-04 23:41:43 -07:00
Roy Yang
d79f0c6b39 Simplify the logic by removing dead code and enhance logging
Signed-off-by: Roy Yang <royyang@google.com>
2020-06-01 23:43:04 -07:00
Kubernetes Prow Robot
5592b5d67a Merge pull request #91470 from MHBauer/fail-0-remote-images
explicitly fail if no images are found when running remote tests
2020-06-01 20:58:14 -07:00
Kubernetes Prow Robot
46d08c89ab Merge pull request #91363 from alejandrox1/tune-node-perf-workloads
Tuned npb is workload resources
2020-05-30 23:25:53 -07:00
Kubernetes Prow Robot
6b15b1f4a6 Merge pull request #91467 from bobbypage/topology-manager-test
Mark Topology Manager Test as non-alpha and NodeFeature
2020-05-26 16:49:14 -07:00
Kubernetes Prow Robot
30eeacbf22 Merge pull request #91384 from alejandrox1/alejandrox1-patch-1
Added cadvisor test suite to flag info message
2020-05-26 16:48:56 -07:00
Morgan Bauer
58924c2de5 explicitly fail if no images are found when running remote tests
The previous implementation succeeds if no images are run. This causes
silent failures when image matchers are provided that do not match any image.
2020-05-26 14:08:27 -07:00
David Porter
f5b8c3d746 Mark Topology Manager Test as non-alpha and NodeFeature 2020-05-26 12:10:18 -07:00
Jorge Alarcon Ochoa
a069eec2bb Added cadvisor test suite to flag info message
The cadvisor test suite is not mentioned in the remote runner's
`--test-suite` flag.
This PR will mention the existence of the cadvisor test suite.
2020-05-23 19:10:32 -04:00
Stephen Augustus
b692502a9d Update CNI to v0.8.6
Signed-off-by: Stephen Augustus <saugustus@vmware.com>
2020-05-22 17:48:56 -04:00
alejandrox1
ebd84a5517 Tuned npb is workload resources
Lowering the amount of cpu allocated to this workload will set the
resources allocated to be similar to the other npb and tf workload in
this tests.
This will also allow to run all three workloads in a n1-standard-12 gcp
instance - which has 16 cpus and 60 GB.

Signed-off-by: alejandrox1 <alarcj137@gmail.com>
2020-05-22 09:30:43 -04:00
Davanum Srinivas
0608e8be25 update bazel BUILD files
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
2020-05-20 10:57:47 -04:00
Davanum Srinivas
5692926914 Move packages for slightly better UX for consumers
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
2020-05-20 10:57:46 -04:00
Davanum Srinivas
07d88617e5 Run hack/update-vendor.sh
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
2020-05-16 07:54:33 -04:00
Davanum Srinivas
442a69c3bd switch over k/k to use klog v2
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
2020-05-16 07:54:27 -04:00
Giuseppe Scrivano
b21b1a5436 test, e2e_node: drop superfluous systemd properties
commit 43c56eb403 introduced a change
where CPUAccounting, CPUAccounting and TasksAccounting are enabled for
the systemd service.

It causes a regression on RHEL 7.8 where systemd-run doesn't allow to
set TasksAccounting.

Since Delegate= already enables all the controllers, it is superfluous
to specify them.

Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-05-04 09:22:35 +02:00
Kubernetes Prow Robot
a62cfe8451 Merge pull request #75111 from tnozicka/fix-e2e-watches
Fix watches in e2e tests
2020-04-23 19:20:07 -07:00
Tomas Nozicka
c62db98e95 Update Bazel 2020-04-23 17:27:13 +02:00
Tomas Nozicka
d0a4c52392 Fix watches in apparmor e2e 2020-04-23 17:26:28 +02:00
tanjunchen
de99aaf8d2 test/e2e_node/gpu_device_plugin_test.go:Remove prometheus dependencies from k/k 2020-04-23 14:23:43 +08:00
Kubernetes Prow Robot
d92fdebd85 Merge pull request #89897 from giuseppe/test-e2e-node
kubelet: fix e2e-node cgroups test on cgroup v2
2020-04-20 15:54:12 -07:00
Ed Bartosh
88478f3749 e2e_node: check if image exists locally before pulling
'docker pull' is a time consuming operation. It makes sense to check
if image exists locally before pulling it from a registry.

Checked if image exists by running 'docker inspect'. Only pull if
image doesn't exist.
2020-04-20 12:27:29 +03:00
tanjunchen
f76da50c7d test/e2e/framework/util.go:move DsFromManifest to test/e2e/framework/manifest , and rename it to DaemonSetFromURL 2020-04-14 09:54:41 +08:00
Giuseppe Scrivano
43c56eb403 e2e_node: adapt tests to cgroup v2
and fix node_container_manager_test to run with the systemd cgroup
manager.

Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
2020-04-09 16:18:05 +02:00
Andrew Sy Kim
2e56866c97 move apparmor annotation constants to k8s.io/api/core/v1
Signed-off-by: Andrew Sy Kim <kim.andrewsy@gmail.com>
2020-04-06 10:22:04 -04:00
Kubernetes Prow Robot
4e9dd8fd36 Merge pull request #89454 from gavinfish/import-aliases
Update .import-aliases for e2e test framework
2020-03-27 14:35:54 -07:00
张潇
a6d0f8e3dc beta.kubernetes.io/arch is already deprecated 2020-03-25 13:34:43 +08:00
drfish
dfab6b637f Update .import-aliases for e2e test framework 2020-03-25 11:40:02 +08:00
Kubernetes Prow Robot
89dfebb214 Merge pull request #89359 from gongguan/process
eviction by process number
2020-03-24 15:27:25 -07:00
louisgong
0efb70c0a2 eviction by process number 2020-03-24 09:25:04 +08:00
tanjunchen
bed22fbb44 WaitForPodReady is simply wrapper functions for e2epod package,
and they made an invalid dependency to sub e2e framework from the core framework.

So we can use e2epod.WaitTimeoutForPodReadyInNamespace to remove invalid dependency.

The main purpose of this pr is to handle the framework core package dependency subpackage pod.
2020-03-22 23:08:52 +08:00
Odin Ugedal
a233b9aab0 Add verbose message when more than one kubelet is running 2020-03-19 13:08:08 +01:00
Odin Ugedal
8b6160a367 Add support for stopping kubelet in node-e2e
This makes it possible to stop the kubelet, do some work, and then
start it again.
2020-03-19 13:08:08 +01:00
Odin Ugedal
2830827442 Add support for removing unsupported huge page sizes
When kubelet is restarted, it will now remove the resources for huge
page sizes no longer supported. This is required when:
- node disables huge pages
- changing the default huge page size in older versions of linux
(because it will then only support the newly set default).
- Software updates that change what sizes are supported (eg. by changing
boot parameters).
2020-03-19 13:08:08 +01:00
Kubernetes Prow Robot
5708511499 Merge pull request #88708 from mikedanese/deleteopts
Migrate clientset metav1.DeleteOpts to pass-by-value
2020-03-05 23:09:23 -08:00
Kubernetes Prow Robot
50dd75f9c5 Merge pull request #88773 from vpickard/e2e-topology-manager-sriovdpReady
e2e-topology-manager: Wait for SR-IOV device plugin
2020-03-05 20:04:38 -08:00
Kubernetes Prow Robot
e23e7204f2 Merge pull request #88558 from egernst/e2e_node-PodOverhead
e2e node pod overhead
2020-03-05 20:04:11 -08:00
Mike Danese
76f8594378 more artisanal fixes
Most of these could have been refactored automatically but it wouldn't
have been uglier. The unsophisticated tooling left lots of unnecessary
struct -> pointer -> struct transitions.
2020-03-05 14:59:47 -08:00
Mike Danese
aaf855c1e6 deref all calls to metav1.NewDeleteOptions that are passed to clients.
This is gross but because NewDeleteOptions is used by various parts of
storage that still pass around pointers, the return type can't be
changed without significant refactoring within the apiserver. I think
this would be good to cleanup, but I want to minimize apiserver side
changes as much as possible in the client signature refactor.
2020-03-05 14:59:46 -08:00
Mike Danese
c58e69ec79 automated refactor 2020-03-05 14:59:46 -08:00
Kubernetes Prow Robot
1f2e1967d1 Merge pull request #88566 from Deepthidharwar/topology-mgr-numa-tests
Enable running cpu-mgr-multiNUMA e2e tests with Topology manager
2020-03-05 05:38:37 -08:00
vpickard
61565b3f6c e2e-topology-manager: Wait for SR-IOV device plugin
Make sure the SR-IOV device plugin is ready, and that
there are enough SR-IOV devices allocatable before
spinning up test pods.

Signed-off-by: vpickard <vpickard@redhat.com>
2020-03-04 10:07:35 -05:00
Deepthi Dharwar
1ede096465 Enable topology-manager-e2e tests to run on MultiNUMA nodes.
Signed-off-by: Deepthi Dharwar <ddharwar@redhat.com>
2020-03-02 22:36:43 +05:30
Deepthi Dharwar
4abbce4549 Refactor CPUMananger-e2e-tests so that it be reused by topology-manager-e2e-testsuite.
Signed-off-by: Deepthi Dharwar <ddharwar@redhat.com>
2020-03-02 22:36:31 +05:30
Deepthi Dharwar
a4b59a5d7c Currently SRIOV detection logic is reporting error if it fails to detect SRIOV device
on the system. This patch aims to fix the same.

Signed-off-by: Deepthi Dharwar <ddharwar@redhat.com>
2020-03-02 19:31:37 +05:30
Eric Ernst
aa12e1f8c4 e2e_node add test for PodOverhead feature
This test will verify that the Pod cgroup created takes Overhead into
account.

Signed-off-by: Eric Ernst <eric@amperecomputing.com>
2020-02-28 23:00:39 +00:00
Kubernetes Prow Robot
624da8b9a3 Merge pull request #88110 from fromanirh/refactor-get-current-kubelet-conf
e2e: e2e_node: refactor getCurrentKubeletConfig
2020-02-24 13:11:36 -08:00