Switching to helm installation is not working

[santurini]

I had a contour installation done via YAML and a bunch of applications using the contour ingress class.

kubectl apply -f https://projectcontour.io/quickstart/contour.yaml

cat <<EOF | kubectl apply -f -
  apiVersion: networking.k8s.io/v1
  kind: IngressClass
  metadata:
    annotations:
      ingressclass.kubernetes.io/is-default-class: "true"
      meta.helm.sh/release-name: contour
      meta.helm.sh/release-namespace: projectcontour
    labels:
      app.kubernetes.io/component: contour
      app.kubernetes.io/instance: contour
      app.kubernetes.io/managed-by: Helm
      app.kubernetes.io/name: contour
      io.portainer.kubernetes.application.name: contour
      io.portainer.kubernetes.application.owner: admin
    name: contour
  spec:
    controller: projectcontour.io/projectcontour/contour-contour
EOF

I wanted to switch to the helm installation so I deleted the namespace, ingress class and all other resources:

kubectl delete -f https://projectcontour.io/quickstart/contour.yaml
kubectl delete ingressclass contour

And reinstalled contour via helm following the official documentation, and I also checked to have everything up and running, ingressclass included.
But I am no more able to reach the applications hosts, while reverting back to the YAML installation I solve the error.

I am using version contour-20.0.3, while in another cluster I did the first installation via helm with version contour-17.0.9 and it works perfectly fine but there are some differences in the pods deployed:

NAME                           READY   STATUS    RESTARTS       AGE
pod/contour-55676fcf76-6rwxq   1/1     Running   0              74d
pod/contour-55676fcf76-lgdwg   1/1     Running   0              49d
pod/contour-envoy-9npd7        2/2     Running   2 (106d ago)   111d
pod/contour-envoy-dmpzq        2/2     Running   2 (50d ago)    67d
pod/contour-envoy-k6w74        2/2     Running   4 (105d ago)   112d
pod/contour-envoy-qwpzg        2/2     Running   2 (106d ago)   111d
pod/envoy-cl2zg                2/2     Running   2 (106d ago)   111d
pod/envoy-fsftf                2/2     Running   2 (50d ago)    67d
pod/envoy-qhckt                2/2     Running   2 (106d ago)   111d
pod/envoy-qtrmt                2/2     Running   4 (105d ago)   112d

NAME                    TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)                      AGE
service/contour         ClusterIP      10.43.181.185   <none>        8001/TCP                     418d
service/contour-envoy   LoadBalancer   10.43.59.74     <pending>     80:31691/TCP,443:30669/TCP   418d
service/envoy           LoadBalancer   10.43.124.25    <pending>     80:30997/TCP,443:30167/TCP   408d

In the new version I only have contour and contour-envoy pods and services, while the envoy is missing and is not mentioned also in the documentation, maybe is that the problem?

[sunjayBhatia]

Looks like you may not have deleted your envoy DaemonSet from the YAML install?

[santurini]

On the node where it is working is a fresh helm install, without YAML.
On the node where helm is not working, I've deleted all the resources and the namespace so it should also be a fresh install.