Add ingress, change root path in deployment

kubernetes/deployment.yaml

@@ -30,6 +30,8 @@ spec:
         - name: flagged-data
           mountPath: /demo/flagged
         env:
+          - name: APP_PATH
+            value: /compress
           - name: LLM_ENDPOINT
             valueFrom:
               secretKeyRef:

kubernetes/ingress.yaml

@@ -0,0 +1,24 @@
+apiVersion: networking.k8s.io/v1
+kind: Ingress
+metadata:
+  name: llmlingua-demo-ingress
+  namespace: cornelius-ns
+  annotations:
+    nginx.ingress.kubernetes.io/rewrite-target: /
+spec:
+  ingressClassName: nginx
+  tls:
+  - hosts:
+    - misc.llm.scads.ai
+    secretName: misc.llm.scads.ai-tls
+  rules:
+  - host: misc.llm.scads.ai
+    http:
+      paths:
+      - path: /compress
+        pathType: Prefix
+        backend:
+          service:
+            name: llmlingua-demo-service
+            port:
+              number: 80

kubernetes/service.yaml

@@ -1,14 +1,12 @@
 apiVersion: v1
 kind: Service
 metadata:
-  name: llmlingua-demo-nodeport
+  name: llmlingua-demo-service
   namespace: cornelius-ns
 spec:
-  type: NodePort
   ports:
     - port: 80
       targetPort: 8000
-      nodePort: 30007
       protocol: TCP
   selector:
     app: llmlingua-demo

src/app.py

@@ -37,6 +37,7 @@
 start_load = time.time()
 load_dotenv()
 
+APP_PATH = os.getenv("APP_PATH") or "/"
 LLM_ENDPOINT = os.getenv("LLM_ENDPOINT")
 LLM_TOKEN = os.getenv("LLM_TOKEN")
 LLM_LIST = [
@@ -362,5 +363,5 @@ def flag(prompt, context, compr_prompt, rate, metrics, res_a_obj, res_b_obj, fla
     flag_b.click(flag, inputs=FLAG_COMPONENTS + [flag_b], outputs=[flag_a, flag_n, flag_b], preprocess=False)
 
 
-app = gr.mount_gradio_app(app, demo, path="/")
+app = gr.mount_gradio_app(app, demo, path=APP_PATH)
 print(f"Ready! Loaded in {time.time() - start_load:.2f}s")