dcgm-exporter

---
# Source: dcgm-exporter/templates/serviceaccount.yaml
apiVersion: v1
kind: ServiceAccount
metadata:
  name: dcgm-dcgm-exporter
  labels:
    helm.sh/chart: dcgm-exporter-1.0.1
    app.kubernetes.io/name: dcgm-exporter
    app.kubernetes.io/instance: dcgm
    app.kubernetes.io/version: "1.7.2"
    app.kubernetes.io/managed-by: Tiller
    app.kubernetes.io/component: "dcgm-exporter"
  annotations:
    prometheus.io/scrape: "true"
---

apiVersion: v1
kind: Service
metadata:
  name: dcgm-dcgm-exporter
  labels:
    helm.sh/chart: dcgm-exporter-1.0.1
    app.kubernetes.io/name: dcgm-exporter
    app.kubernetes.io/instance: dcgm
    app.kubernetes.io/version: "1.7.2"
    app.kubernetes.io/managed-by: Tiller
    app.kubernetes.io/component: "dcgm-exporter"
spec:
  type: ClusterIP
  ports:
  - name: "metrics"
    port: 9400
    targetPort: 9400
    protocol: TCP
  selector:
    app.kubernetes.io/name: dcgm-exporter
    app.kubernetes.io/instance: dcgm
---

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: dcgm-dcgm-exporter
  labels:
    helm.sh/chart: dcgm-exporter-1.0.1
    app.kubernetes.io/name: dcgm-exporter
    app.kubernetes.io/instance: dcgm
    app.kubernetes.io/version: "1.7.2"
    app.kubernetes.io/managed-by: Tiller
    app.kubernetes.io/component: "dcgm-exporter"
spec:
  updateStrategy:
    type: RollingUpdate
  selector:
    matchLabels:
      app.kubernetes.io/name: dcgm-exporter
      app.kubernetes.io/instance: dcgm
      app.kubernetes.io/component: "dcgm-exporter"
spec:
  updateStrategy:
    type: RollingUpdate
  selector:
    matchLabels:
      app.kubernetes.io/name: dcgm-exporter
      app.kubernetes.io/instance: dcgm
      app.kubernetes.io/component: "dcgm-exporter"
  template:
    metadata:
      labels:
        app.kubernetes.io/name: dcgm-exporter
        app.kubernetes.io/instance: dcgm
        app.kubernetes.io/component: "dcgm-exporter"
    spec:
      serviceAccountName: dcgm-dcgm-exporter
      nodeSelector:
        node.kubernetes.io/type.gpu: "true"

      volumes:
      - name: "pod-gpu-resources"
        hostPath:
          path: "/var/lib/kubelet/pod-resources"
      containers:
      - name: exporter
        securityContext:
          runAsNonRoot: false
          runAsUser: 0

        image: "nvidia/dcgm-exporter:1.7.2"
        imagePullPolicy: "IfNotPresent"
        env:
        - name: "DCGM_EXPORTER_KUBERNETES"
          value: "true"
        - name: "DCGM_EXPORTER_LISTEN"
          value: "9400"
        ports:
        - name: "metrics"
          containerPort: 9400
        volumeMounts:
        - name: "pod-gpu-resources"
          readOnly: true
          mountPath: "/var/lib/kubelet/pod-resources"
        livenessProbe:
          httpGet:
            path: /health
            port: 9400
          initialDelaySeconds: 5
          periodSeconds: 5
        readinessProbe:
          httpGet:
            path: /health
            port: 9400
          initialDelaySeconds: 5
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: dcgm-dcgm-exporter
  labels:
    helm.sh/chart: dcgm-exporter-1.0.1
    app.kubernetes.io/name: dcgm-exporter
    app.kubernetes.io/instance: dcgm
    app.kubernetes.io/version: "1.7.2"
    app.kubernetes.io/managed-by: Tiller
    app.kubernetes.io/component: "dcgm-exporter"
    release: genauto-k8s-prom

spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: dcgm-exporter
      app.kubernetes.io/instance: dcgm
      app.kubernetes.io/component: "dcgm-exporter"
      release: "genauto-k8s-prom"
  endpoints:
  - port: "metrics"
    path: "/metrics"
  namespaceSelector:
    matchNames:
    - "monitoring"
  endpoints:
  - port: "metrics"
    path: "/metrics"
    interval: "15s"