lidarr ollama ui, soulseek

roles/cripto/files/ai-ollama-deployment.yaml

@@ -9,6 +9,93 @@ spec:
 ---
 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
@@ -30,9 +117,16 @@ spec:
         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
@@ -62,7 +156,7 @@ metadata:
   namespace: fenix-ai
 spec:
   capacity:
-    storage: 20Gi
+    storage: 40Gi
   storageClassName: ollama-ai-nfs-csi
   accessModes:
     - ReadWriteOnce
@@ -82,7 +176,7 @@ spec:
     - ReadWriteOnce
   resources:
     requests:
-      storage: 20Gi
+      storage: 40Gi
 ---
 apiVersion: storage.k8s.io/v1
 kind: StorageClass