titan-iac/services/monitoring/dcgm-exporter.yaml

# services/monitoring/dcgm-exporter.yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: dcgm-exporter
  namespace: monitoring
  labels:
    app: dcgm-exporter
spec:
  selector:
    matchLabels:
      app: dcgm-exporter
  updateStrategy:
    rollingUpdate:
      maxUnavailable: 2
  template:
    metadata:
      labels:
        app: dcgm-exporter
      annotations:
        prometheus.io/scrape: "true"
        prometheus.io/port: "9400"
    spec:
      serviceAccountName: default
      runtimeClassName: nvidia
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
              - matchExpressions:
                  - key: kubernetes.io/hostname
                    operator: In
                    values:
                      - titan-20
                      - titan-21
                      - titan-22
                      - titan-24
      tolerations:
        - operator: Exists
      containers:
        - name: dcgm-exporter
          image: registry.bstein.dev/monitoring/dcgm-exporter:4.4.2-4.7.0-ubuntu22.04
          imagePullPolicy: Always
          ports:
            - name: metrics
              containerPort: 9400
          env:
            - name: DCGM_EXPORTER_KUBERNETES
              value: "true"
          securityContext:
            privileged: true
          resources:
            requests:
              cpu: 50m
              memory: 64Mi
          volumeMounts:
            - name: pod-resources
              mountPath: /var/lib/kubelet/pod-resources
      imagePullSecrets:
        - name: zot-regcred
      volumes:
        - name: pod-resources
          hostPath:
            path: /var/lib/kubelet/pod-resources
            type: Directory
---
apiVersion: v1
kind: Service
metadata:
  name: dcgm-exporter
  namespace: monitoring
  labels:
    app: dcgm-exporter
spec:
  selector:
    app: dcgm-exporter
  ports:
    - name: metrics
      port: 9400
      targetPort: metrics