fix ?

manifests/metrics-server/configmap.yaml

@@ -1,53 +1,90 @@
 apiVersion: v1
 kind: ConfigMap
 metadata:
-  name: argocd-metrics-server-configmap
+  name: argocd-metrics-server
 data:
-  # Metrics for Deployments
+  # Prometheus URL - ADJUST THIS to match your Prometheus service
-  extension.metrics.deployments: |
-    - name: "CPU Usage"
-      description: "CPU usage for the deployment"
-      type: "graph"
-      graphType: "area"
-      yAxisLabel: "CPU Cores"
-      query: 'sum(rate(container_cpu_usage_seconds_total{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}[5m])) by (pod)'
-
-    - name: "Memory Usage"
-      description: "Memory usage for the deployment"
-      type: "graph"
-      graphType: "area"
-      yAxisLabel: "Memory (MB)"
-      query: 'sum(container_memory_working_set_bytes{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}) by (pod) / 1024 / 1024'
-
-    - name: "Network Received"
-      description: "Network bytes received"
-      type: "graph"
-      graphType: "line"
-      yAxisLabel: "Bytes/sec"
-      query: 'sum(rate(container_network_receive_bytes_total{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}[5m])) by (pod)'
-
-  # Metrics for Pods
-  extension.metrics.pods: |
-    - name: "CPU Usage"
-      description: "CPU usage for the pod"
-      type: "graph"
-      graphType: "area"
-      yAxisLabel: "CPU Cores"
-      query: 'sum(rate(container_cpu_usage_seconds_total{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}[5m])) by (container)'
-
-    - name: "Memory Usage"
-      description: "Memory usage for the pod"
-      type: "graph"
-      graphType: "area"
-      yAxisLabel: "Memory (MB)"
-      query: 'sum(container_memory_working_set_bytes{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}) by (container) / 1024 / 1024'
-
-    - name: "Container Restarts"
-      description: "Number of container restarts"
-      type: "graph"
-      graphType: "line"
-      yAxisLabel: "Restarts"
-      query: 'kube_pod_container_status_restarts_total{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}'
-
-  # Prometheus connection
   prometheus.url: "http://prometheus-kube-prometheus-prometheus.monitoring.svc:9090"
+  
+  # Metrics configuration for Deployments
+  extension.metrics.deployments: |
+    metrics:
+      - name: CPU Usage
+        description: Pod CPU usage  
+        graphType: area
+        query: |
+          sum(
+            irate(
+              container_cpu_usage_seconds_total{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}[2m]
+            )
+          ) by (pod)
+      - name: Memory Usage
+        description: Pod memory usage
+        graphType: area
+        query: |
+          sum(
+            container_memory_working_set_bytes{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}
+          ) by (pod)
+      - name: Replicas
+        description: Deployment replicas
+        graphType: line
+        query: |
+          sum(
+            kube_deployment_status_replicas{namespace="{{.metadata.namespace}}", deployment="{{.metadata.name}}"}
+          )
+
+  # Metrics configuration for Pods
+  extension.metrics.pods: |
+    metrics:
+      - name: CPU Usage
+        description: CPU usage by container
+        graphType: area
+        query: |
+          sum(
+            irate(
+              container_cpu_usage_seconds_total{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}[2m]
+            )
+          ) by (container)
+      - name: Memory Usage
+        description: Memory usage by container
+        graphType: area
+        query: |
+          sum(
+            container_memory_working_set_bytes{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}
+          ) by (container)
+      - name: Network I/O
+        description: Network traffic
+        graphType: line
+        query: |
+          sum(
+            rate(
+              container_network_transmit_bytes_total{namespace="{{.metadata.namespace}}", pod="{{.metadata.name}}"}[2m]
+            )
+          ) by (pod)
+
+  # Metrics configuration for Argo Rollouts
+  extension.metrics.rollouts: |
+    metrics:
+      - name: CPU Usage
+        description: Pod CPU usage
+        graphType: area
+        query: |
+          sum(
+            irate(
+              container_cpu_usage_seconds_total{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}[2m]
+            )
+          ) by (pod)
+      - name: Memory Usage
+        description: Pod memory usage
+        graphType: area
+        query: |
+          sum(
+            container_memory_working_set_bytes{namespace="{{.metadata.namespace}}", pod=~"{{.metadata.name}}-.*"}
+          ) by (pod)
+      - name: Replicas
+        description: Rollout replicas
+        graphType: line
+        query: |
+          sum(
+            kube_deployment_status_replicas{namespace="{{.metadata.namespace}}", deployment=~"{{.metadata.name}}-.*"}
+          )

manifests/metrics-server/deployment.yaml

@@ -3,7 +3,8 @@ kind: Service
 metadata:
   name: argocd-metrics-server
   labels:
-    app: argocd-metrics-server
+    app.kubernetes.io/name: argocd-metrics-server
+    app.kubernetes.io/part-of: argocd
 spec:
   ports:
   - name: http
@@ -11,41 +12,44 @@ spec:
     protocol: TCP
     targetPort: 9003
   selector:
-    app: argocd-metrics-server
+    app.kubernetes.io/name: argocd-metrics-server
 ---
 apiVersion: apps/v1
 kind: Deployment
 metadata:
   name: argocd-metrics-server
   labels:
-    app: argocd-metrics-server
+    app.kubernetes.io/name: argocd-metrics-server
+    app.kubernetes.io/part-of: argocd
+    app.kubernetes.io/component: metrics-server
 spec:
   replicas: 1
   selector:
     matchLabels:
-      app: argocd-metrics-server
+      app.kubernetes.io/name: argocd-metrics-server
   template:
     metadata:
       labels:
-        app: argocd-metrics-server
+        app.kubernetes.io/name: argocd-metrics-server
     spec:
+      serviceAccountName: argocd-server
       containers:
       - name: argocd-metrics-server
         image: quay.io/argoprojlabs/argocd-extension-metrics:v1.0.3
         imagePullPolicy: IfNotPresent
+        command:
+        - /app/argocd-extension-metrics
+        - backend
+        - start
+        - --port=9003
+        - --config=/etc/config/config.yaml
         ports:
         - containerPort: 9003
           name: http
           protocol: TCP
-        env:
-        - name: PROMETHEUS_URL
-          valueFrom:
-            configMapKeyRef:
-              name: argocd-metrics-server-configmap
-              key: prometheus.url
         volumeMounts:
         - name: config
-          mountPath: /etc/argocd-metrics-server/
+          mountPath: /etc/config
         resources:
           limits:
             cpu: 100m
@@ -60,14 +64,14 @@ spec:
             - ALL
           readOnlyRootFilesystem: true
           runAsNonRoot: true
-          runAsUser: 1000
+          runAsUser: 999
       volumes:
       - name: config
         configMap:
-          name: argocd-metrics-server-configmap
+          name: argocd-metrics-server
       securityContext:
         runAsNonRoot: true
-        runAsUser: 1000
+        runAsUser: 999
-        fsGroup: 1000
+        fsGroup: 999
         seccompProfile:
           type: RuntimeDefault

manifests/metrics-server/kustomization.yaml

@@ -8,6 +8,6 @@ resources:
   - deployment.yaml
 
 commonLabels:
-  app: argocd-metrics-server
   app.kubernetes.io/name: argocd-metrics-server
+  app.kubernetes.io/part-of: argocd
   app.kubernetes.io/component: metrics-server