79599ac419
Prune internal clientset/informer from kubeapiserver admission initializer ( #70167 )
...
* externalize pv resize admission controller
* externalize podtolerationrestriction admission controller
* externalize podnodeselector admission controller
* remove internal clientset/informer from kubeapiserver admission initializer
* minor change: fixes scheduler integration test compiliation
2018-10-24 14:47:16 -07:00
10121e6448
Merge pull request #69898 from Huang-Wei/scheudler-perf-more-cases
...
add more scheduler benchmark testcases
2018-10-24 14:47:06 -07:00
86e2d2bf03
Merge pull request #69587 from jonfriesen/fix_golint_tests
...
Fixes golint for test/list
2018-10-24 14:46:57 -07:00
5259d09c38
add more scheduler benchmark testcases
...
- add benchmark test for PodAffinity
- add benchmark test for NodeAffinity
- add 1000-nodes test for PodAntiAffinity/PodAffinity/NodeAffinity
2018-10-24 11:06:32 -07:00
38ddb4413a
update token authn constructor
2018-10-24 16:56:38 +08:00
daf17e2aee
Merge pull request #69571 from bclau/tests-linux-commands-fix
...
tests: Makes tests' commands platform agnostic
2018-10-23 09:33:30 -07:00
8be25aaade
Merge pull request #70125 from jbartosik/e2e-deflake
...
Deflake e2e tests of HPA
2018-10-23 08:21:30 -07:00
83d3f7033b
Merge pull request #69441 from mkimuram/provision
...
Move minimum set of dynamic provisioning e2e test to testsuites
2018-10-23 07:08:35 -07:00
e2394bd3b8
Deflake e2e tests of HPA
...
Resource consumer might use slightly more CPU than requested. That
resulted in HPA sometimes increasing size of deployments during e2e
tests. Deflake tests by:
- Scaling up CPU requests in those tests. Resource consumer might go a fixed
number of milli CPU seconds above target. Having higher requests makes
the test less sensitive.
- On scale down consume CPU in the middle between what would generate
recommendation of expexted size and 1 pod fewer (instead of righ on
edge beween expected and expected +1).
Some variables were int32 but always cast to int before use. Make them
int.
2018-10-23 14:50:41 +02:00
fecaa546c9
Merge pull request #69593 from dekkagaijin/legacy-metadata-test
...
account for disabled legacy metadata endpoints
2018-10-23 03:20:44 -07:00
b1fbdfe76b
Merge pull request #69993 from Pradip-Khakurel/issue-#34059-annotate-auth-e2e-errors
...
make error messages more helpful for some e2e auth tests
2018-10-23 02:08:22 -07:00
4d182cec03
Merge pull request #70105 from mikedanese/trev1
...
promote --service-account-api-audiences to top level kube-apiserver config
2018-10-22 23:59:34 -07:00
baa8d800ad
Merge pull request #69582 from mikedanese/trev7
...
tokenreview: authenticator interface changes
2018-10-22 23:59:04 -07:00
2472d67357
Merge pull request #68218 from jcsirot/fix-SkipIfMultizone-message
...
Fix the skip message since it was the exact opposite of the tested co…
2018-10-22 22:46:32 -07:00
77a1199b6c
Merge pull request #70019 from enj/enj/t/shared_etcd_stubs
...
Refactor dry run test to reuse ETCD storage data
2018-10-22 19:44:29 -07:00
371b1e7fed
promote --service-account-api-audiences to top level kube-apiserver config
...
The service account authenticator isn't the only authenticator that
should respect API audience. The authentication config structure should
reflect that.
2018-10-22 18:21:37 -07:00
e5227216c0
rebase authenticators onto new interface.
2018-10-22 10:16:59 -07:00
4fcb36e205
Move minimum set of dynamic provisioning e2e test to testsuites
2018-10-22 16:04:43 +00:00
570f9195d0
Merge pull request #69829 from jpbetz/watch-e2e-test1
...
Add api-machinery 'watch-consistency' e2e test
2018-10-19 22:17:54 -07:00
3787a4b5be
register skeleton provider
2018-10-19 17:05:06 -07:00
82cba09870
Merge pull request #69995 from ixdy/purge-libsystemd-from-debian-base
...
debian-base: purge libsystemd0 to eliminate CVE false-positives
2018-10-19 16:04:23 -07:00
c16edb2738
Refactor dry run test to reuse ETCD storage data
...
This change updates the ETCD storage test so that its data is
exported. Thus it can be used by other tests. The dry run test was
updated to consume this data instead of having a duplicate copy.
The code to start a master that can be used for "one of every
resource" style tests was also factored out. It is reused in the
dry run test as well.
This prevents these tests from drifting in the future and reduces
the long term maintenance burden.
Signed-off-by: Monis Khan <mkhan@redhat.com>
2018-10-19 18:35:56 -04:00
f87d2c61f3
Add api-machinery 'watch-consistency' e2e test
2018-10-19 14:05:43 -07:00
8b36038b41
Merge pull request #68483 from pohly/e2e-refactor-pr
...
e2e refactor
2018-10-19 12:32:01 -07:00
7a8696c331
Update to debian-base 0.4.0
...
Additionally, update the addon-manager to use kubectl v1.11.3.
2018-10-19 11:13:28 -07:00
c0d9a0728c
Merge pull request #69897 from code-sleuth/develop
...
Remove error output from stdout to stderr
2018-10-19 04:57:33 -07:00
cfd6343477
Merge pull request #66515 from justinsb/verify_kubectl_dump
...
Add test for kubectl cluster-info dump
2018-10-18 18:38:52 -07:00
c30f5269a3
Merge pull request #69810 from chrisohaver/checkmem
...
Add DNS container memory check to scalability tests
2018-10-18 14:44:09 -07:00
7480650e05
Remove error output from stdout to stderr
2018-10-19 00:28:36 +03:00
8c4fd312c9
make error messages more helpful for some e2e auth tests
2018-10-18 22:03:02 +02:00
402266afde
add dns pod memory constraint
2018-10-18 10:35:30 -04:00
0ab29b4852
Merge pull request #69944 from jingxu97/defertest
...
Make sure kubelet is restarted after it is stopped in the test
2018-10-18 00:10:54 -07:00
0652e098d0
Merge pull request #67547 from pbarker/audit-api
...
dynamic audit configuration api
2018-10-17 16:19:39 -07:00
9de354034b
Make sure kubelet is restarted in the storage test
...
In some storage tests, kubelet is stopped first and the test check node
NotReady state. However, if it fails to have this state, kubelet could
not be restarted because the defer function is placed after the stop
kubelet command. This PR fixes this issue.
2018-10-17 14:19:04 -07:00
897b3a9fa7
Merge pull request #69437 from ravisantoshgudimetla/promote-resource-limits-priority-function
...
Promote resource limits priority function to beta
2018-10-17 11:57:02 -07:00
bba6e2583e
Merge pull request #69739 from mortent/UnflakeCmdRsTest
...
Remove flaky test by avoiding reuse of name for replicaset
2018-10-16 19:26:05 -07:00
1d3d1e8968
Merge pull request #68518 from leakingtapan/golint-fix-volume-empty-dir
...
fix golint issues for pkg/volume/empty_dir
2018-10-16 17:52:44 -07:00
851444f382
Merge pull request #68300 from yliaog/e2e-api
...
Upgraded e2e/apimachinery/aggregator to 1.10 Sample API server
2018-10-16 16:23:17 -07:00
08be38186d
Merge pull request #69755 from mrunalp/use_fsync
...
Use fsync instead of sync for test
2018-10-16 13:01:52 -07:00
dbd3be9213
Merge pull request #67695 from ipuustin/start-kubemark
...
start-kubemark.sh: clean up the shell script.
2018-10-16 10:03:04 -07:00
ba71ae7227
E2E and build file generated
2018-10-16 12:58:38 -04:00
b8e1250487
adds dynamic audit api generated
2018-10-16 13:07:27 +00:00
381d0a5d14
adds dynamic audit api
2018-10-16 06:46:34 -06:00
8726ed29ef
enable and improve e2e of NoExecuteTaintManager
...
- e2e test file cannot end with "_test.go". rename "taints_test.go" to "taints.go"
- test multiple pods case
2018-10-15 16:02:05 -07:00
ce1e62fb1b
Upgraded the test to 1.10 Sample API server.
2018-10-15 15:35:24 -07:00
c7886f8ba4
Merge pull request #69828 from ironcladlou/e2e-psp-reentrancy-fix
...
Make CreatePrivilegedPSPBinding reentrant
2018-10-15 15:04:53 -07:00
6327c7403b
Use fsync instead of sync for test
...
sync sometimes hangs with misbehaving NFS during the tests
Signed-off-by: Mrunal Patel <mpatel@redhat.com>
2018-10-15 14:27:26 -07:00
d54e0fc0bc
Merge pull request #69670 from atlassian/simplify-rand-seed
...
Simplify rand seeding
2018-10-15 13:53:01 -07:00
42ef7ae8d5
fix golint issues for pkg/volume/empty_dir
2018-10-15 20:00:36 +00:00
0bef5e4325
Make CreatePrivilegedPSPBinding reentrant
...
Make CreatePrivilegedPSPBinding reentrant so tests using it (e.g. DNS) can be
executed more than once against a cluster. Without this change, such tests will
fail because the PSP already exists, short circuiting test setup.
2018-10-15 15:31:25 -04:00
Kim Min