# NVIDIA Device Plugin v0.20.1 (Quelle: deployments/static/nvidia-device-plugin.yml des Projekts),
# ergänzt um runtimeClassName nvidia und nodeSelector knaster.gpu=true (nur die GPU-VMs).
# Meldet je Knoten die Ressource nvidia.com/gpu, die Pods anfordern können.
apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: nvidia-device-plugin-daemonset
  namespace: kube-system
spec:
  selector:
    matchLabels:
      name: nvidia-device-plugin-ds
  updateStrategy:
    type: RollingUpdate
  template:
    metadata:
      labels:
        name: nvidia-device-plugin-ds
    spec:
      runtimeClassName: nvidia
      nodeSelector:
        knaster.gpu: "true"
      tolerations:
        - key: nvidia.com/gpu
          operator: Exists
          effect: NoSchedule
      priorityClassName: system-node-critical
      containers:
        - image: nvcr.io/nvidia/k8s-device-plugin:v0.20.1
          name: nvidia-device-plugin-ctr
          env:
            - name: FAIL_ON_INIT_ERROR
              value: "false"
            # Nur Index 0 (Quadro M4000 in allen GPU-VMs, geprüft in playbooks/gpu-llamacpp-k3s.yml) wird als
            # nvidia.com/gpu vergeben. Die P2000 gehört fest den llama.cpp-Pods (per UUID, Task 6.1).
            - name: NVIDIA_VISIBLE_DEVICES
              value: "0"
          securityContext:
            allowPrivilegeEscalation: false
            capabilities:
              drop: ["ALL"]
          volumeMounts:
            - name: device-plugin
              mountPath: /var/lib/kubelet/device-plugins
      volumes:
        - name: device-plugin
          hostPath:
            path: /var/lib/kubelet/device-plugins