Configure Prometheus to access DCGM metrics

myeolenv · Mar 16, 2020 · 2062173 · 2062173
1 parent f916504
commit 2062173
Show file tree

Hide file tree

Showing 7 changed files with 46 additions and 0 deletions.
diff --git a/assets/state-monitoring/0400_prom_role.yaml b/assets/state-monitoring/0400_prom_role.yaml
@@ -0,0 +1,16 @@
+apiVersion: rbac.authorization.k8s.io/v1
+kind: Role
+metadata:
+  name: prometheus-k8s
+  namespace: gpu-operator-resources
+rules:
+- apiGroups:
+  - ""
+  resources:
+  - services
+  - endpoints
+  - pods
+  verbs:
+  - get
+  - list
+  - watch
diff --git a/assets/state-monitoring/0500_prom_rolebinding.yaml b/assets/state-monitoring/0500_prom_rolebinding.yaml
@@ -0,0 +1,13 @@
+apiVersion: rbac.authorization.k8s.io/v1
+kind: RoleBinding
+metadata:
+  name: prometheus-k8s
+  namespace: gpu-operator-resources
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: Role
+  name: prometheus-k8s
+subjects:
+- kind: ServiceAccount
+  name: prometheus-k8s
+  namespace: openshift-monitoring
diff --git a/assets/state-monitoring/0400_service.yaml → assets/state-monitoring/0600_service.yaml b/assets/state-monitoring/0400_service.yaml → assets/state-monitoring/0600_service.yaml
diff --git a/assets/state-monitoring/0700_service_monitor.yaml b/assets/state-monitoring/0700_service_monitor.yaml
@@ -0,0 +1,15 @@
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  name: nvidia-dcgm-exporter
+  namespace: gpu-operator-resources
+spec:
+  endpoints:
+  - port: gpu-metrics
+  jobLabel: app
+  namespaceSelector:
+    matchNames:
+    - gpu-operator-resources
+  selector:
+    matchLabels:
+     app: nvidia-dcgm-exporter
diff --git a/assets/state-monitoring/0500_configmap.yaml → assets/state-monitoring/0800_configmap.yaml b/assets/state-monitoring/0500_configmap.yaml → assets/state-monitoring/0800_configmap.yaml
diff --git a/assets/state-monitoring/0600_daemonset.yaml → assets/state-monitoring/0900_daemonset.yaml b/assets/state-monitoring/0600_daemonset.yaml → assets/state-monitoring/0900_daemonset.yaml
diff --git a/deployments/gpu-operator/templates/resources-namespace.yaml b/deployments/gpu-operator/templates/resources-namespace.yaml
@@ -2,3 +2,5 @@ apiVersion: v1
 kind: Namespace
 metadata:
     name: gpu-operator-resources
+    labels:
+      openshift.io/cluster-monitoring: "true"