You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First off, thank you for creating and maintaining this awesome project!
After applying policy-controller on our infra namespaces we realized a kind of Chicken-Egg problem with CoreDNS and Policy-controller. The problem is that the policy-controller's service name cannot be resolved without a CoreDNS pod running, but the CoreDNS replicasets and pods cannot start, since they can't be mutated because the service which the MutatingWebhook calls cannot be resolved. Of course, we temporarily removed the flag from the namespace added it back after the pods have started, and rollout restarted everything but this will come up as a problem again in any cluster shutdown or disaster recovery scenario.
Do you have any suggestions to solve this somehow?
Thank you in advance!
The text was updated successfully, but these errors were encountered:
Description
First off, thank you for creating and maintaining this awesome project!
After applying policy-controller on our infra namespaces we realized a kind of Chicken-Egg problem with CoreDNS and Policy-controller. The problem is that the policy-controller's service name cannot be resolved without a CoreDNS pod running, but the CoreDNS replicasets and pods cannot start, since they can't be mutated because the service which the MutatingWebhook calls cannot be resolved. Of course, we temporarily removed the flag from the namespace added it back after the pods have started, and rollout restarted everything but this will come up as a problem again in any cluster shutdown or disaster recovery scenario.
Do you have any suggestions to solve this somehow?
Thank you in advance!
The text was updated successfully, but these errors were encountered: