try to fix this broken prometheus in staging

k8s/monitoring/values-staging.yaml

@@ -1,38 +1,43 @@
 # Helm chart values for Prometheus Operator with HTTPS and basic auth
+# Explicitly enable RBAC resource creation
+rbac:
+  create: true
+
+
 kube-prometheus-stack:
-  nodeExporter:
+
-    enabled: true
+  # Disable the default ServiceMonitor configuration paths to prevent duplicates
+  prometheus-node-exporter:
     serviceMonitor:
-      enabled: true
+      enabled: false
-      relabelings:
+  nodeExporter:
-      - sourceLabels: [__meta_kubernetes_pod_node_name]
+    serviceMonitor:
-        targetLabel: instance
+      enabled: false
-        action: replace
+
-      # Optional Rule 2: If the original IP address needs to be retained
-      # as a separate label (e.g., 'ip_address'), this rule can be added:
-      - sourceLabels: [__address__]
-        targetLabel: ip_address
-        action: replace
   prometheus:
     ingress:
       enabled: true
+      pathType: Prefix
       annotations:
         kubernetes.io/ingress.class: nginx
-        nginx.ingress.kubernetes.io/rewrite-target: /$2
+        #nginx.ingress.kubernetes.io/rewrite-target: /$2
         cert-manager.io/cluster-issuer: letsencrypt-nginx-production
-        # nginx.ingress.kubernetes.io/auth-type: basic
+        nginx.ingress.kubernetes.io/backend-protocol: "HTTP"
-        # nginx.ingress.kubernetes.io/auth-secret: basic-auth
+        nginx.ingress.kubernetes.io/auth-type: basic
-        # nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
+        nginx.ingress.kubernetes.io/auth-secret: monitoring-basic-auth 
+        nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
       hosts:
       - monitoring.staging.video.jamkazam.com
       paths:
-      - /prometheus(/|$)(.*)
+      - /prometheus
       tls:
       - secretName: monitoring
         hosts:
         - monitoring.staging.video.jamkazam.com
     prometheusSpec:
-      routePrefix: /
+      retention: 60d
+      retentionSize: 20GB
+      routePrefix: /prometheus
       externalUrl: https://monitoring.staging.video.jamkazam.com/prometheus
       storageSpec:
         volumeClaimTemplate:
@@ -40,28 +45,83 @@ kube-prometheus-stack:
             storageClassName: linode-block-storage-retain
             resources:
               requests:
-                storage: 10Gi
+                storage: 30Gi
+      # 2. !!! CRUCIAL: Ensure the default ServiceMonitor is ignored !!!
+      # This prevents duplicate metrics by telling Prometheus to ignore the default SM.
+      serviceMonitorSelector:
+        matchExpressions:
+          # Exclude the default node-exporter ServiceMonitor
+          - key: app.kubernetes.io/name
+            operator: NotIn
+            values:
+              # Use the label identified above
+              - prometheus-node-exporter
+      serviceMonitorNamespaceSelector:
+        matchExpressions:
+          - key: kubernetes.io/metadata.name
+            operator: In
+            values:
+              - monitoring  # Its own namespace
+              - webrtc-be   # Your app's namespace
+      # Add the manual scrape configuration
+      additionalScrapeConfigs:
+        - job_name: 'node-exporter'
+          kubernetes_sd_configs:
+            - role: endpoints
+          
+          relabel_configs:
+            # 1. Filter: Precisely target the node-exporter service in the monitoring namespace.
+            - source_labels: [__meta_kubernetes_namespace, __meta_kubernetes_service_name]
+              separator: '/'
+              # Assuming the service name is 'monitoring-prometheus-node-exporter'
+              regex: 'monitoring/monitoring-prometheus-node-exporter'
+              action: keep
+            
+            # 2. Filter: Ensure we are targeting the standard port (usually 9100)
+            - source_labels: [__address__]
+              regex: '.*:9100$' 
+              action: keep
 
+            # 3. THE FIX: Set the instance label correctly
+            - source_labels: [__meta_kubernetes_endpoint_node_name]
+              target_label: instance
+              action: replace
+            - source_labels: [__address__]
+              target_label: ip_address
+              action: replace
+
+            # 4. Replicate standard labels for dashboard compatibility
+            - action: labelmap
+              regex: __meta_kubernetes_pod_label_(.+)
+            # Ensure standard labels are present for dashboard compatibility
+            - source_labels: [__meta_kubernetes_namespace]
+              target_label: namespace
+            - source_labels: [__meta_kubernetes_pod_name]
+              target_label: pod
+            - source_labels: [__meta_kubernetes_endpoint_node_name]
+              target_label: node
   alertmanager:
     ingress:
       enabled: true
+      pathType: Prefix
       annotations:
         kubernetes.io/ingress.class: nginx
-        nginx.ingress.kubernetes.io/rewrite-target: /$2
+        #nginx.ingress.kubernetes.io/rewrite-target: /$2
         cert-manager.io/cluster-issuer: letsencrypt-nginx-production
-        # nginx.ingress.kubernetes.io/auth-type: basic
+        nginx.ingress.kubernetes.io/backend-protocol: "HTTP"
-        # nginx.ingress.kubernetes.io/auth-secret: basic-auth
+        nginx.ingress.kubernetes.io/auth-type: basic
-        # nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
+        nginx.ingress.kubernetes.io/auth-secret: monitoring-basic-auth 
+        nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
       hosts:
       - monitoring.staging.video.jamkazam.com
       paths:
-      - /alertmanager(/|$)(.*)
+      - /alertmanager
       tls:
       - secretName: monitoring
         hosts:
         - monitoring.staging.video.jamkazam.com
     alertmanagerSpec:
-      routePrefix: /
+      routePrefix: /alertmanager
       externalUrl: https://monitoring.staging.video.jamkazam.com/alertmanager
       storage:
         volumeClaimTemplate:
@@ -69,25 +129,27 @@ kube-prometheus-stack:
             storageClassName: linode-block-storage-retain
             resources:
               requests:
-                storage: 10Gi
+                storage: 30Gi
 
   grafana:
     persistence:
       enabled: true
       storageClassName: linode-block-storage-retain
-      size: 10Gi
+      size: 30Gi
     ingress:
       enabled: true
+      pathType: Prefix
       annotations:
         kubernetes.io/ingress.class: nginx
-        nginx.ingress.kubernetes.io/rewrite-target: /$2
+        #nginx.ingress.kubernetes.io/rewrite-target: /$2
         cert-manager.io/cluster-issuer: letsencrypt-nginx-production
-        # nginx.ingress.kubernetes.io/auth-type: basic
+        nginx.ingress.kubernetes.io/backend-protocol: "HTTP"
-        # nginx.ingress.kubernetes.io/auth-secret: basic-auth
+        #nginx.ingress.kubernetes.io/auth-type: basic
-        # nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
+        #nginx.ingress.kubernetes.io/auth-secret: monitoring-basic-auth 
+        #nginx.ingress.kubernetes.io/auth-realm: 'Authentication Required'
       hosts:
       - monitoring.staging.video.jamkazam.com
-      path: /grafana(/|$)(.*)
+      path: /grafana
       tls:
       - secretName: monitoring
         hosts:
@@ -97,6 +159,7 @@ kube-prometheus-stack:
         domain: monitoring.staging.video.jamkazam.com
         root_url: "%(protocol)s://%(domain)s/grafana/"
         enable_gzip: "true"
+        serve_from_sub_path: true
 
   # Disable control plane metrics
   kubeEtcd:
@@ -107,4 +170,4 @@ kube-prometheus-stack:
 
   kubeScheduler:
     enabled: false
-      
+