Hi, we have installed the last versione of the nginx-ingress-controller, but if we follow the documentation specificaly for DO, and set the proxy-protocol, this are the configurations applied:
controller: service: type: LoadBalancer externalTrafficPolicy: Local annotations: service.beta.kubernetes.io/do-loadbalancer-enable-proxy-protocol: "true" kubernetes.digitalocean.com/load-balancer-id: XXXXXXXXXXXXXX config: use-proxy-protocol: "true" admissionWebhooks: timeoutSeconds: 29 metrics: enabled: true service: annotations: prometheus.io/scrape: "true" prometheus.io/port: "10254"
But with this configurations we see a bunch of errors on the ingress with truncated headers and from inside of any container if we make a curl with https to a domain present in the cluster, we see:
GnuTLS: Error in the pull function
By the way if we issue a new certificate with certmanager, the certmanager can’t close the loop with the normal http01
I know that it’s not much logs that i give, but now the cluster is in production and the configurations for proxy-protocol are disable so that we don’t have errors. If the errors are necessary I will re-deploy the configurations to get the specific errors.
PS: we have migrated from an old ingress deployment to a newone and that because there is the load-balancer-id