Closed jbartosik closed 2 years ago
When I run e2e tests in my cluster from current master in logs of API server I see logs like:
"failed calling webhook "vpa.k8s.io": expected webhook response of admission.k8s.io/v1, Kind=AdmissionReview, got /, Kind="
What do you have configured for admissionReviewVersions
in vpa webhook in your cluster? Was the VPA webhook manually created and vpa admission-controller started with--register-webhook=False
?
Is there a way I can run the E2E tests suite locally, on my laptop?
Is there a way I can run the E2E tests suite locally, on my laptop?
I don't know. I create a cluster and use ./hack/deploy-for-e2e.sh admission-controller
to deploy admission controller from what's in my repo, then ./hack/run-e2e-tests.sh admission-controller
to run e2e test.
What do you have configured for
admissionReviewVersions
in vpa webhook in your cluster? Was the VPA webhook manually created and vpa admission-controller started with--register-webhook=False
?
$ kubectl get MutatingWebhookConfiguration vpa-webhook-config -oyaml
apiVersion: admissionregistration.k8s.io/v1
kind: MutatingWebhookConfiguration
metadata:
creationTimestamp: {{scrubbed}}
generation: 1
name: vpa-webhook-config
resourceVersion: {{scrubbed}}
uid: {{scrubbed}}
webhooks:
- admissionReviewVersions:
- v1
clientConfig:
caBundle: {{scrubbed}}
service:
name: vpa-webhook
namespace: kube-system
port: 443
failurePolicy: Ignore
matchPolicy: Equivalent
name: vpa.k8s.io
namespaceSelector: {}
objectSelector: {}
reinvocationPolicy: Never
rules:
- apiGroups:
- ""
apiVersions:
- v1
operations:
- CREATE
resources:
- pods
scope: '*'
- apiGroups:
- autoscaling.k8s.io
apiVersions:
- '*'
operations:
- CREATE
- UPDATE
resources:
- verticalpodautoscalers
scope: '*'
sideEffects: None
timeoutSeconds: 30
VPA e2e tests are consistently failing. They were consistently passing (flake once every few days) until recently.
For
full
suite:For
admission-controller
:For
actuation
:It looks like admission-controller is broken. I suspect that the culprit is #4537