New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Resource Quota test with uninitialized pod fails #72742

Open
bclau opened this Issue Jan 9, 2019 · 6 comments

Comments

Projects
None yet
6 participants
@bclau
Copy link
Contributor

bclau commented Jan 9, 2019

Which jobs are failing:

Unknown.

Which test(s) are failing:

[Feature:Initializers] should create a ResourceQuota and capture the life of an uninitialized pod.

Since when has it been failing:

Unknown.

Testgrid link:

Unknown

Reason for failure:

INFO: expect err to be timeout error, got Pod "too-large-pod" is invalid: metadata.initializers.pending[0].name: Invalid value: "unhandled": should be a domain with at least three segments separated by dots

Even after that particular issue is fixed, the test continues to fail at the step when creating a Pod with an Initializer which exceeds the Resource Quota:

INFO: expect err to be timeout error, got pods "too-large-pod" is forbidden: exceeded quota: t est-quota, requested: cpu=1100m, used: cpu=0, limited: cpu=1

Additionally, in my case, the 30 seconds wait for the Resource Quota to be updated is not always enough, which causes the test to sometimes timeout.

Anything else we need to know:

/sig testing

@bclau bclau referenced a pull request that will close this issue Jan 9, 2019

Open

tests: Fixes resource quota related test #72743

@krmayankk

This comment has been minimized.

Copy link
Contributor

krmayankk commented Jan 10, 2019

@bclau which test is failing ? is this from PR or a UT or integration ?

@bclau

This comment has been minimized.

Copy link
Contributor

bclau commented Jan 10, 2019

@krmayankk The test is mentioned in the description: [Feature:Initializers] should create a ResourceQuota and capture the life of an uninitialized pod., which is an e2e test.

@k8s-ci-robot k8s-ci-robot added sig/testing and removed needs-sig labels Jan 10, 2019

@bsalamat

This comment has been minimized.

Copy link
Contributor

bsalamat commented Jan 10, 2019

@bsalamat

This comment has been minimized.

Copy link
Contributor

bsalamat commented Jan 11, 2019

Some of the tests also fail in other setups, such as:
https://k8s-testgrid.appspot.com/sig-release-1.12-all#gke-cos-1.12-default
https://k8s-testgrid.appspot.com/sig-release-1.13-all#gke-cos-1.13-default

In all the failures, logs similar to the ones below are seen:

STEP: Building a namespace api object, basename resourcequota
STEP: Waiting for a default service account to be provisioned in namespace
[It] should create a ResourceQuota and capture the life of a configMap.
  /go/src/k8s.io/kubernetes/_output/dockerized/go/src/k8s.io/kubernetes/test/e2e/scheduling/resource_quota.go:367
STEP: Creating a ResourceQuota
STEP: Ensuring resource quota status is calculated
Jan 10 22:28:28.246: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:29.851: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:31.851: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:33.854: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:35.863: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:38.008: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:39.933: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:41.940: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:43.856: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:45.892: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:47.850: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:49.851: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:51.850: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:53.966: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:55.851: INFO: resource resourcequotas, expected 1, actual 2
Jan 10 22:28:55.891: INFO: resource resourcequotas, expected 1, actual 2
[AfterEach] [sig-scheduling] ResourceQuota

It looks like there are changes in new versions of the clusters that breaks the assumptions of the tests.

@krmayankk

This comment has been minimized.

Copy link
Contributor

krmayankk commented Jan 11, 2019

@bclau i meant where is the source code for these tests found it here test/e2e/apimachinery/resource_quota.go

@fedebongio

This comment has been minimized.

Copy link
Contributor

fedebongio commented Jan 14, 2019

/assign @yliaog

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment