job.yaml

{{$HostNetworkMode := DefaultParam .CL2_USE_HOST_NETWORK_PODS false}}
{{$RUN_ON_ARM_NODES := DefaultParam .CL2_RUN_ON_ARM_NODES false}}
{{$TOLERATION := DefaultParam .CL2_TOLERATION ""}}
{{$Image := DefaultParam .Image "registry.k8s.io/pause:3.9"}}
{{$podPayloadSize := DefaultParam .CL2_JOB_POD_PAYLOAD_SIZE 0}}
{{$RuntimeClassName := DefaultParam .CL2_RUNTIME_CLASS_NAME nil}}

apiVersion: batch/v1
kind: Job
metadata:
  name: {{.Name}}
  labels:
    group: load
spec:
  backoffLimit: {{.ReplicasMin}}
  manualSelector: true
  parallelism: {{RandIntRange .ReplicasMin .ReplicasMax}}
  completions: {{.Completions}}
  selector:
    matchLabels:
      name: {{.Name}}
  template:
    metadata:
      labels:
        group: load
        name: {{.Name}}
    spec:
      {{if $RuntimeClassName}}
      runtimeClassName: {{$RuntimeClassName}}
      {{end}}
      hostNetwork: {{$HostNetworkMode}}
      containers:
      - name: {{.Name}}
        # TODO(#799): We should test the "run-to-completion" workflow and hence don't use pause pods.
        image: {{$Image}}
        env:
        - name: ENV_VAR
          value: {{RandDataWithSeed $podPayloadSize .Name}}
        resources:
          # Keep the CpuRequest/MemoryRequest request equal percentage of 1-core, 4GB node.
          # For now we're setting it to 0.5%.
          requests:
            cpu: 5m
            memory: "20M"
      restartPolicy: OnFailure
      terminationGracePeriodSeconds: 1
      # Add not-ready/unreachable tolerations for 15 minutes so that node
      # failure doesn't trigger pod deletion.
      tolerations:
        - key: "node.kubernetes.io/not-ready"
          operator: "Exists"
          effect: "NoExecute"
          tolerationSeconds: 900
        - key: "node.kubernetes.io/unreachable"
          operator: "Exists"
          effect: "NoExecute"
          tolerationSeconds: 900
        {{if $RUN_ON_ARM_NODES}}
        - key: "kubernetes.io/arch"
          operator: Equal
          value: arm64
          effect: NoSchedule
        {{end}}
        {{if $TOLERATION}}
        - key: {{$TOLERATION}}
          operator: Exists
          effect: NoSchedule
        {{end}}