tenant-catalog/kubeflow/experimental/ray/raycluster_example.yaml

apiVersion: security.istio.io/v1beta1
kind: AuthorizationPolicy
metadata:
  name: allow-ray-workers-head
spec:
  action: ALLOW
  rules:
  - from:
    - source:
        principals:
        # kubeflow-user-example-com should be replaced with the namespace where the Ray cluster is being deployed
        # TODO automatically use the current namespace
        - "cluster.local/ns/kubeflow-user-example-com/sa/default-editor"
  - to:
    - operation:
        ports:
        - "6379"
        - "6380"
        - "6381"
        - "6382"
        - "6383"
        - "52365"
        - "8080"
        - "10012"
---
apiVersion: v1
kind: Service
metadata:
  labels:
    ray.io/headless-worker-svc: raycluster-istio
  name: raycluster-istio-headless-svc
spec:
  clusterIP: None
  selector:
    ray.io/cluster: kubeflow-raycluster
  publishNotReadyAddresses: true
  ports:
  - name: node-manager-port
    port: 6380
    appProtocol: grpc
  - name: object-manager-port
    port: 6381
    appProtocol: grpc
  - name: runtime-env-agent-port
    port: 6382
    appProtocol: grpc
  - name: dashboard-agent-grpc-port
    port: 6383
    appProtocol: grpc
  - name: dashboard-agent-listen-port
    port: 52365
    appProtocol: http
  - name: metrics-export-port
    port: 8080
    appProtocol: http
  - name: max-worker-port
    port: 10012
    appProtocol: grpc
---
apiVersion: ray.io/v1
kind: RayCluster
metadata:
  name: kubeflow-raycluster
spec:
  rayVersion: '2.44.1'
  enableInTreeAutoscaling: true
  autoscalerOptions:
    upscalingMode: Default 
    idleTimeoutSeconds: 60
  headGroupSpec:
    rayStartParams:
      num-cpus: '1'
      node-manager-port: '6380'
      object-manager-port: '6381'
      runtime-env-agent-port: '6382'
      dashboard-agent-grpc-port: '6383'
      dashboard-agent-listen-port: '52365'
      metrics-export-port: '8080'
      max-worker-port: '10012'
      # kubeflow-user-example-com should be replaced with the namespace where the Ray cluster is being deployed
      # TODO automatically use the current namespace
      node-ip-address: $(hostname -I | tr -d ' ' | sed 's/\./-/g').raycluster-istio-headless-svc.kubeflow-user-example-com.svc.cluster.local
    template:
      metadata:
        labels:
          sidecar.istio.io/inject: "true"
      spec:
        serviceAccountName: default-editor
        containers:
        - name: ray-head # TODO why call it headless with a head...
          image: rayproject/ray:2.44.1-py311-cpu
          lifecycle:
            preStop:
              exec:
                command: ["/bin/sh","-c","ray stop"]
          volumeMounts:
            - mountPath: /tmp/ray
              name: ray-logs
          resources:
            limits:
              cpu: "1"
              memory: "2G"
            requests:
              cpu: "100m"
              memory: "2G"
          securityContext:
            allowPrivilegeEscalation: false
            capabilities:
              drop: ["ALL"]
            runAsNonRoot: true
            seccompProfile:
              type: RuntimeDefault
        volumes:
          - name: ray-logs
            emptyDir: {}
  workerGroupSpecs:
    - replicas: 1
      minReplicas: 1
      maxReplicas: 1
      groupName: small-group
      rayStartParams:
        num-cpus: '1'
        node-manager-port: '6380'
        object-manager-port: '6381'
        runtime-env-agent-port: '6382'
        dashboard-agent-grpc-port: '6383'
        dashboard-agent-listen-port: '52365'
        metrics-export-port: '8080'
        max-worker-port: '10012'
        # kubeflow-user-example-com should be replaced with the namespace where the Ray cluster is being deployed
        # TODO automatically use the current namespace
        node-ip-address: $(hostname -I | tr -d ' ' | sed 's/\./-/g').raycluster-istio-headless-svc.kubeflow-user-example-com.svc.cluster.local
      template:
        metadata:
          labels:
            sidecar.istio.io/inject: "true"
        spec:
          serviceAccountName: default-editor
          containers:
          - name: ray-worker
            image: rayproject/ray:2.44.1-py311-cpu
            lifecycle:
              preStop:
                exec:
                  command: ["/bin/sh","-c","ray stop"]
            # use volumeMounts.Optional.
            # Refer to https://kubernetes.io/docs/concepts/storage/volumes/
            volumeMounts:
             - mountPath: /tmp/ray
               name: ray-logs
            resources:
             limits:
               cpu: "1"
               memory: "1G"
             requests:
               cpu: "300m"
               memory: "1G"
            securityContext:
             allowPrivilegeEscalation: false
             capabilities:
               drop: ["ALL"]
             runAsNonRoot: true
             seccompProfile:
               type: RuntimeDefault
          volumes:
            - name: ray-logs
              emptyDir: {}