add environment-specific cilium profiles

local-quickstart/cilium.md

@@ -16,22 +16,26 @@ Kubernetes service VIP because kube-proxy is disabled during bootstrap.
 ```bash
 kubectl config current-context
 kubectl get nodes -o wide
-cp deployments/cilium/cilium-values.yaml /tmp/cilium-values.yaml
-# Replace REPLACE_WITH_API_SERVER_DNS_OR_PRIVATE_IP with the reachable API endpoint.
+# The profile is composed from common settings and the k3d overlay.
+# Replace the endpoint values in a temporary file if installing manually.
+helm_values_dir=deployments/cilium/values
 helm repo add cilium https://helm.cilium.io/
 helm repo update
 helm upgrade --install cilium cilium/cilium \
   --namespace kube-system \
   --version 1.20.0 \
-  --values /tmp/cilium-values.yaml
+  --values "$helm_values_dir/common.yaml" \
+  --values "$helm_values_dir/profiles/k3d.yaml" \
+  --set k8sServiceHost=REPLACE_WITH_API_SERVER_DNS_OR_PRIVATE_IP \
+  --set k8sServicePort=REPLACE_WITH_API_SERVER_PORT
 kubectl -n kube-system rollout status daemonset/cilium --timeout=300s
 kubectl -n kube-system get pods -l k8s-app=cilium
 ```
 
 Install ArgoCD only after Cilium is Ready. Apply
-`deployments/argocd/optional-apps/cilium.yaml` manually after the endpoint is configured;
-it is intentionally outside the root app-of-apps watched directory until the
-cluster has been bootstrapped without kube-proxy.
+`deployments/argocd/optional-apps/cilium-k3d.yaml` manually only after the
+endpoint is configured. It is intentionally outside the root app-of-apps
+watched directory until the cluster has been bootstrapped without kube-proxy.
 
 Delete the profile with: