lab-deployment.yaml

kind: Deployment
apiVersion: apps/v1
metadata:
  name: fine-tuned-merlinite-7b
  labels:
    app: fine-tuned-merlinite-7b
spec:
  replicas: 1
  selector:
    matchLabels:
      app: fine-tuned-merlinite-7b
  template:
    metadata:
      labels:
        app: fine-tuned-merlinite-7b
    spec:
      restartPolicy: Always
      schedulerName: default-scheduler
      affinity: {}
      terminationGracePeriodSeconds: 120
      securityContext: {}
      containers:
        - resources:
            limits:
              nvidia.com/gpu: '1'
            requests:
              nvidia.com/gpu: '1'
          readinessProbe:
            httpGet:
              path: /health
              port: http
              scheme: HTTP
            timeoutSeconds: 5
            periodSeconds: 30
            successThreshold: 1
            failureThreshold: 3
          terminationMessagePath: /dev/termination-log
          name: server
          livenessProbe:
            httpGet:
              path: /health
              port: http
              scheme: HTTP
            timeoutSeconds: 8
            periodSeconds: 100
            successThreshold: 1
            failureThreshold: 3
          env: []
          args: [
            "--model",
            "/opt/app-root/src/training_results/final",
            "--tokenizer",
            "/opt/app-root/src/training_results/final"
          ]
          securityContext:
            capabilities:
              drop:
                - ALL
            runAsNonRoot: true
            allowPrivilegeEscalation: false
            seccompProfile:
              type: RuntimeDefault
          ports:
            - name: http
              containerPort: 8000
              protocol: TCP
          imagePullPolicy: IfNotPresent
          startupProbe:
            httpGet:
              path: /health
              port: http
              scheme: HTTP
            initialDelaySeconds: 60
            timeoutSeconds: 1
            periodSeconds: 30
            successThreshold: 1
            failureThreshold: 24
          volumeMounts:
            - name: shm
              mountPath: /dev/shm
            - name: instruct-lab-pvc
              mountPath: /opt/app-root/src
          terminationMessagePolicy: File
          image: 'quay.io/rh-aiservices-bu/vllm-openai-ubi9:0.3.3'
      volumes:
        - name: shm
          emptyDir:
            medium: Memory
            sizeLimit: 1Gi
        - name: instruct-lab-pvc
          persistentVolumeClaim:
            claimName: lab-train-shared-data-pvc
      dnsPolicy: ClusterFirst
      tolerations:
        - key: tenant
          value: rh-jarvis-ai
          effect: NoSchedule
  strategy:
    type: RollingUpdate
    rollingUpdate:
      maxUnavailable: 25%
      maxSurge: 1
  revisionHistoryLimit: 10
  progressDeadlineSeconds: 600

---

kind: Service
apiVersion: v1
metadata:
  name: fine-tuned-merlinite-7b
  labels:
    app: fine-tuned-merlinite-7b
spec:
  clusterIP: None
  ipFamilies:
    - IPv4
  ports:
    - name: http
      protocol: TCP
      port: 8000
      targetPort: http
  type: ClusterIP
  ipFamilyPolicy: SingleStack
  sessionAffinity: None
  selector:
    app: fine-tuned-merlinite-7b

---

kind: Route
apiVersion: route.openshift.io/v1
metadata:
  name: fine-tuned-merlinite-7b
  labels:
    shard: internal
    app: fine-tuned-merlinite-7b
spec:
  host: fine-tuned-merlinite-7b.apps.int.stc.ai.dev.us-east-1.aws.paas.redhat.com
  to:
    kind: Service
    name: fine-tuned-merlinite-7b
    weight: 100
  port:
    targetPort: http
  tls:
    termination: edge