Closed flavio closed 2 months ago
This was caused by the missing finalizer. This is not solved inside of the main
branch of the helm chart (which creates the (Validating|Mutating)WebhookConfiguration
of our CRDs.
This is going to be part of 1.17.0-RC4
Is there an existing issue for this?
Current Behavior
Testing Kubewarden 1.17.0-rc3
I've created a
ClusterAdmissionPolicyGroup
policy, which lead to the creation of aValidatingWebhookConfiguration
resource.Then I removed the
ClusterAdmissionPolicyGroup
. The configuration of the Policy Server was updated accordingly, but the associatedValidatingWebhookConfiguration
was not removed.This is a big deal, the webhook will still be registered, but unreachable. Hence all the request relevant for this webhook will fail. This can cause a DOS inside of the cluster.
Expected Behavior
The
`ValidatingWebhookConfiguration
resource is removed immediately, even before the configuration of the Policy Server is updated.Steps To Reproduce
Deploy kubewarden 1.17.0-rc3.
Create the following policy:
Wait for the policy to be active. Get the list of
ValidatingWebhookConfiguration
resources:Remove the policy:
Get the list of
ValidatingWebhookConfiguration
resources. Regardless of the amount of time waited, this is the output:Environment
Anything else?
No response