I have encountered situations where the CCO controller Pod is not functioning and regularly crashing, but due to no readinessProbe, looks fine on the surface. This has happened with the CCO in Mint Mode while using an AWS Provider when the kube-system/aws-creds information becomes invalid due to IAM access/secret keys being deactivated after a rotation.
I have previously seen messages in the logs indicating AWS permission issues, but lately the CCO container logs are silent. The only symptom was never-ending messages like these in the CCO Pod's kube-rbac-proxy container's logs:
I0519 19:48:43.671759 1 log.go:245] http: proxy error: dial tcp 127.0.0.1:2112: connect: connection refused
cloud-credential-operator/manifests/03-deployment.yaml
Line 34 in 6a880b4
I have encountered situations where the CCO controller Pod is not functioning and regularly crashing, but due to no readinessProbe, looks fine on the surface. This has happened with the CCO in Mint Mode while using an AWS Provider when the kube-system/aws-creds information becomes invalid due to IAM access/secret keys being deactivated after a rotation.
I have previously seen messages in the logs indicating AWS permission issues, but lately the CCO container logs are silent. The only symptom was never-ending messages like these in the CCO Pod's kube-rbac-proxy container's logs: