feat: enable Traefik ServiceMonitor in deploy-monitoring.sh and update deploy-stack.sh for improved observability

Author: KevinDeBenedetti

kubernetes/ingress/traefik-values.yaml

@@ -76,10 +76,10 @@ metrics:
     addEntryPointsLabels: true
     addRoutersLabels: true
     addServicesLabels: true
-    # ServiceMonitor: auto-discovered by kube-prometheus-stack's Prometheus Operator
+    # ServiceMonitor: enabled by deploy-monitoring.sh once kube-prometheus-stack CRDs exist
     disableAPICheck: true
     serviceMonitor:
-      enabled: true
+      enabled: false
       namespace: ingress
       jobLabel: traefik
       interval: 30s

scripts/deploy-monitoring.sh

@@ -80,6 +80,18 @@ kubectl rollout status deployment/kube-prometheus-stack-grafana -n monitoring --
 log_info "Waiting for Prometheus to be ready..."
 kubectl rollout status statefulset/prometheus-kube-prometheus-stack-prometheus -n monitoring --timeout=120s
 
+# Enable Traefik ServiceMonitor now that kube-prometheus-stack CRDs exist
+log_step "Enabling Traefik ServiceMonitor..."
+helm upgrade traefik traefik/traefik \
+  --namespace ingress \
+  --reuse-values \
+  --set "metrics.prometheus.serviceMonitor.enabled=true" \
+  --set "metrics.prometheus.serviceMonitor.namespace=ingress" \
+  --set "metrics.prometheus.serviceMonitor.jobLabel=traefik" \
+  --set "metrics.prometheus.serviceMonitor.interval=30s" \
+  --wait \
+  --timeout 60s
+
 # --- 3. Loki ---
 log_step "[3/5] Loki ${LOKI_VERSION}..."
 helm upgrade --install loki grafana/loki \

scripts/deploy-stack.sh

@@ -41,14 +41,8 @@ log_info "Deploying base stack on cluster: $(kubectl config current-context)"
 log_step "[1/4] Namespaces..."
 kubectl apply -f "$(_k8s namespaces/namespaces.yaml)"
 
-# --- 2. Prometheus Operator CRDs (required by Traefik ServiceMonitor) ---
-log_step "[2/5] Prometheus Operator CRDs..."
-PROM_OPERATOR_VERSION="${PROM_OPERATOR_VERSION:-v0.82.0}"
-kubectl apply --server-side -f \
-  "https://github.com/prometheus-operator/prometheus-operator/releases/download/${PROM_OPERATOR_VERSION}/stripped-down-crds.yaml"
-
-# --- 3. Traefik (Ingress Controller) ---
-log_step "[3/5] Traefik ${TRAEFIK_CHART_VERSION}..."
+# --- 2. Traefik (Ingress Controller) ---
+log_step "[2/4] Traefik ${TRAEFIK_CHART_VERSION}..."
 helm repo add traefik https://helm.traefik.io/traefik --force-update
 helm repo update traefik
 # Kubernetes forbids loopback addresses as externalIPs; skip the flag for local Lima testing.
@@ -70,8 +64,8 @@ helm upgrade --install traefik traefik/traefik \
 log_info "Waiting for Traefik to be ready..."
 kubectl rollout status deployment/traefik -n ingress --timeout=120s
 
-# --- 4. cert-manager ---
-log_step "[4/5] cert-manager ${CERT_MANAGER_VERSION}..."
+# --- 3. cert-manager ---
+log_step "[3/4] cert-manager ${CERT_MANAGER_VERSION}..."
 helm repo add jetstack https://charts.jetstack.io --force-update
 helm repo update jetstack
 helm upgrade --install cert-manager jetstack/cert-manager \
@@ -90,7 +84,7 @@ kubectl rollout status deployment/cert-manager-webhook -n cert-manager --timeout
 sleep 10
 
 # --- 4. ClusterIssuers ---
-log_step "[5/5] ClusterIssuers..."
+log_step "[4/4] ClusterIssuers..."
 envsubst < "$(_k8s_file cert-manager/clusterissuer.yaml)" | kubectl apply -f -
 
 # --- 5. Traefik dashboard (optional) ---