iac-ansible-public/roles/cripto/files/ai-ollama-deployment.yaml

apiVersion: metallb.io/v1beta1
kind: IPAddressPool
metadata:
  name: local-pool-2
  namespace: metallb-system
spec:
  addresses:
    - 192.168.1.100-192.168.1.200
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: ai-openwebui-deployment
  namespace: fenix-ai
spec:
  replicas: 1
  selector:
    matchLabels:
      app: ai-openwebui
  template:
    metadata:
      labels:
        app: ai-openwebui
    spec:
      containers:
      - name: ai-openwebui
        image: ghcr.io/open-webui/open-webui:cuda
        ports:
        - containerPort: 8080
        env:
          - name: WEBUI_HOST
            value: "0.0.0.0"
          - name: OLLAMA_BASE_URL
            value: "http://ollama-api-svc.fenix-ai.svc.cluster.local:11434"
        volumeMounts:
          - name: ai-openwebui-data
            mountPath: /app/backend/data
      volumes:
      - name: ai-openwebui-data
        persistentVolumeClaim:
          claimName: ai-openwebui-data-pvc
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: ai-openwebui-data-pv
  namespace: fenix-ai
spec:
  capacity:
    storage: 40Gi
  storageClassName: ai-openwebui-nfs-csi
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  nfs:
    server: 192.168.1.22
    path: /mnt/fenix-main-nas-pool-0/data/k8s-Volumes/k8s-cluster-iac-deployed/ai-openwebui
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: ai-openwebui-data-pvc
  namespace: fenix-ai
spec:
  storageClassName: ai-openwebui-nfs-csi
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 40Gi
---
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: ai-openwebui-nfs-csi
  namespace: fenix-ai
provisioner: nfs.csi.k8s.io
parameters:
  server: 192.168.1.22
  share: /mnt/fenix-main-nas-pool-0/data/k8s-Volumes/k8s-cluster-iac-deployed/ai-openwebui
allowVolumeExpansion: true
reclaimPolicy: Retain
---
apiVersion: v1
kind: Service
metadata:
  name: ai-openwebui-svc
  namespace: fenix-ai
spec:
  selector:
    app: ai-openwebui
  ports:
  - port: 8080
    targetPort: 8080
  type: ClusterIP   # ou LoadBalancer se tiveres suporte
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: ollama-deployment
  namespace: fenix-ai
spec:
  replicas: 1
  selector:
    matchLabels:
      app: ollama
  template:
    metadata:
      labels:
        app: ollama
    spec:
      containers:
      - name: ollama
        image: ollama/ollama:latest
        ports:
        - containerPort: 11434
        resources:
          limits:
            nvidia.com/gpu: 1   # garante uso da tua RTX 4060 Ti
            memory: 17Gi
        env:
          - name: OLLAMA_HOST
            value: "0.0.0.0"
          - name: NVIDIA_VISIBLE_DEVICES
            value: "all"
          - name: NVIDIA_DRIVER_CAPABILITIES
            value: "compute,utility"
          - name: CUDA_VISIBLE_DEVICES
            value: "0"
        volumeMounts:
          - name: ollama-data
            mountPath: /root/.ollama
      volumes:
      - name: ollama-data
        persistentVolumeClaim:
          claimName: ollama-data-pvc
---
apiVersion: v1
kind: Service
metadata:
  name: ollama-api-svc
  namespace: fenix-ai
spec:
  selector:
    app: ollama
  ports:
  - protocol: TCP
    port: 11434
    targetPort: 11434
  type: NodePort   # ou LoadBalancer se tiveres suporte
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: ollama-data-pv
  namespace: fenix-ai
spec:
  capacity:
    storage: 40Gi
  storageClassName: ollama-ai-nfs-csi
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  nfs:
    server: 192.168.1.22
    path: /mnt/fenix-main-nas-pool-0/data/k8s-Volumes/k8s-cluster-iac-deployed/ollama
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: ollama-data-pvc
  namespace: fenix-ai
spec:
  storageClassName: ollama-ai-nfs-csi
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 40Gi
---
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: ollama-ai-nfs-csi
  namespace: fenix-ai
provisioner: nfs.csi.k8s.io
parameters:
  server: 192.168.1.22
  share: /mnt/fenix-main-nas-pool-0/data/k8s-Volumes/k8s-cluster-iac-deployed/ollama
allowVolumeExpansion: true
reclaimPolicy: Retain