chore: remove unnecessary proxy layer (#234)

* Remove proxy layer in pages

Signed-off-by: kerthcet <kerthcet@gmail.com>

* Remove proxy layer

Signed-off-by: kerthcet <kerthcet@gmail.com>

---------

Signed-off-by: kerthcet <kerthcet@gmail.com>

Author: kerthcet

alphatrion/server/cmd/main.py

@@ -11,12 +11,11 @@
 from importlib.metadata import PackageNotFoundError, version
 from pathlib import Path
 
-import httpx
 import uvicorn
 from dotenv import load_dotenv
 from faker import Faker
-from fastapi import FastAPI, Request
-from fastapi.responses import FileResponse, Response
+from fastapi import FastAPI
+from fastapi.responses import FileResponse
 from fastapi.staticfiles import StaticFiles
 from rich.console import Console
 from rich.text import Text
@@ -49,19 +48,14 @@ def main():
     server.set_defaults(func=run_server)
 
     dashboard = subparsers.add_parser(
-        "dashboard", help="Launch the AlphaTrion web dashboard"
+        "dashboard",
+        help="Serve pre-built dashboard (for testing production builds locally)",
     )
     dashboard.add_argument(
         "--port",
         type=int,
         default=5173,
-        help="Port to run the dashboard on (default: 5173)",
-    )
-    dashboard.add_argument(
-        "--backend-url",
-        type=str,
-        default="http://localhost:8000",
-        help="Backend server URL to proxy requests to (default: http://localhost:8000)",
+        help="Port to serve the dashboard on (default: 5173)",
     )
     dashboard.set_defaults(func=start_dashboard)
 
@@ -560,99 +554,29 @@ def start_dashboard(args):
     )
     console.print(msg)
     console.print(Text(f"📂 Serving static files from: {static_path}", style="dim"))
-
-    console.print(
-        Text(f"🔗 Proxying backend requests to: {args.backend_url}", style="dim")
-    )
     console.print()
     console.print(
-        Text("💡 Note: Make sure the backend server is running:", style="bold yellow")
+        Text(
+            "💡 Note: Make sure the backend server is running at http://localhost:8000",
+            style="bold yellow",
+        )
     )
-    console.print(Text("   alphatrion server", style="cyan"))
+    console.print(Text("   Run: alphatrion server", style="cyan"))
     console.print()
 
     app = FastAPI()
 
-    # Create HTTP client for proxying requests to backend
-    http_client = httpx.AsyncClient(base_url=args.backend_url, timeout=30.0)
-
-    # Proxy /graphql requests to backend (MUST be before catch-all route)
-    @app.api_route("/graphql", methods=["GET", "POST"])
-    async def proxy_graphql(request: Request):
-        headers = dict(request.headers)
-        headers.pop("host", None)  # Remove host header
-
-        try:
-            response = await http_client.request(
-                method=request.method,
-                url="/graphql",
-                content=await request.body(),
-                headers=headers,
-            )
-            return Response(
-                content=response.content,
-                status_code=response.status_code,
-                headers=dict(response.headers),
-            )
-        except Exception as e:
-            console.print(Text(f"❌ Error connecting to backend: {e}", style="red"))
-            return Response(
-                content=f'{{"error": "Backend server not available. Make sure it\'s running at {args.backend_url}"}}',
-                status_code=503,
-                media_type="application/json",
-            )
-
-    # Proxy /api requests to backend (MUST be before catch-all route)
-    @app.api_route("/api/{path:path}", methods=["GET", "POST", "PUT", "DELETE"])
-    async def proxy_api(path: str, request: Request):
-        headers = dict(request.headers)
-        headers.pop("host", None)
-
-        try:
-            response = await http_client.request(
-                method=request.method,
-                url=f"/api/{path}",
-                content=await request.body(),
-                headers=headers,
-            )
-            return Response(
-                content=response.content,
-                status_code=response.status_code,
-                headers=dict(response.headers),
-            )
-        except Exception as e:
-            console.print(Text(f"❌ Error connecting to backend: {e}", style="red"))
-            return Response(
-                content='{"error": "Backend server not available"}',
-                status_code=503,
-                media_type="application/json",
-            )
-
     # Mount the entire static directory at /static
     app.mount("/static", StaticFiles(directory=static_path, html=True), name="static")
 
-    @app.get("/")
-    def serve_root():
-        # Serve index.html at root
-        index_file = static_path / "index.html"
-        if index_file.exists():
-            return FileResponse(index_file)
-        return {"error": "index.html not found"}
-
+    # SPA fallback: serve index.html for all routes (enables client-side routing)
     @app.get("/{full_path:path}")
     def spa_fallback(full_path: str):
-        # Serve index.html for all routes (SPA fallback)
-        # This enables client-side routing
         index_file = static_path / "index.html"
         if index_file.exists():
             return FileResponse(index_file)
         return {"error": "index.html not found"}
 
-    # Register cleanup handler for HTTP client
-    @app.on_event("shutdown")
-    async def shutdown_event():
-        await http_client.aclose()
-
     url = f"http://127.0.0.1:{args.port}"
 
     console.print(Text(f"🌐 Dashboard URL: {url}", style="bold cyan"))

alphatrion/storage/runtime.py

@@ -75,7 +75,10 @@ def __init__(self):
                 tracer_provider.add_span_processor(CostEnrichmentProcessor())
 
                 # 3. Add Prometheus exporter if enabled
-                if os.getenv(envs.ENABLE_PROMETHEUS_EXPORTER, "false").lower() == "true":
+                if (
+                    os.getenv(envs.ENABLE_PROMETHEUS_EXPORTER, "false").lower()
+                    == "true"
+                ):
                     pushgateway_url = os.getenv(
                         envs.PROMETHEUS_PUSHGATEWAY_URL, "localhost:9091"
                     )

alphatrion/tracing/prometheus_exporter.py

@@ -206,7 +206,14 @@ def _process_span(self, span: ReadableSpan):
                 self.llm_errors_total.labels(error_type=error_type).inc()
 
             self._process_llm_span(
-                span, attributes, org_id, team_id, user_id, experiment_id, duration, status
+                span,
+                attributes,
+                org_id,
+                team_id,
+                user_id,
+                experiment_id,
+                duration,
+                status,
             )
 
         except Exception as e:
@@ -272,7 +279,11 @@ def _process_llm_span(
 
         if input_tokens > 0:
             self.llm_input_tokens_total.labels(
-                org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                org_id=org_id,
+                team_id=team_id,
+                user_id=user_id,
+                experiment_id=experiment_id,
+                model=model,
             ).inc(input_tokens)
             self.llm_tokens_total.labels(
                 org_id=org_id,
@@ -285,7 +296,11 @@ def _process_llm_span(
 
         if output_tokens > 0:
             self.llm_output_tokens_total.labels(
-                org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                org_id=org_id,
+                team_id=team_id,
+                user_id=user_id,
+                experiment_id=experiment_id,
+                model=model,
             ).inc(output_tokens)
             self.llm_tokens_total.labels(
                 org_id=org_id,
@@ -298,7 +313,11 @@ def _process_llm_span(
 
         if cache_read_input_tokens > 0:
             self.llm_cache_read_input_tokens_total.labels(
-                org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                org_id=org_id,
+                team_id=team_id,
+                user_id=user_id,
+                experiment_id=experiment_id,
+                model=model,
             ).inc(cache_read_input_tokens)
             self.llm_tokens_total.labels(
                 org_id=org_id,
@@ -311,7 +330,11 @@ def _process_llm_span(
 
         if cache_creation_input_tokens > 0:
             self.llm_cache_creation_input_tokens_total.labels(
-                org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                org_id=org_id,
+                team_id=team_id,
+                user_id=user_id,
+                experiment_id=experiment_id,
+                model=model,
             ).inc(cache_creation_input_tokens)
             self.llm_tokens_total.labels(
                 org_id=org_id,
@@ -346,35 +369,60 @@ def _process_llm_span(
 
             if input_cost > 0:
                 self.llm_input_cost_total.labels(
-                    org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                    org_id=org_id,
+                    team_id=team_id,
+                    user_id=user_id,
+                    experiment_id=experiment_id,
+                    model=model,
                 ).inc(input_cost)
 
             if output_cost > 0:
                 self.llm_output_cost_total.labels(
-                    org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                    org_id=org_id,
+                    team_id=team_id,
+                    user_id=user_id,
+                    experiment_id=experiment_id,
+                    model=model,
                 ).inc(output_cost)
 
             if cache_read_cost > 0:
                 self.llm_cache_read_cost_total.labels(
-                    org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                    org_id=org_id,
+                    team_id=team_id,
+                    user_id=user_id,
+                    experiment_id=experiment_id,
+                    model=model,
                 ).inc(cache_read_cost)
 
             if cache_creation_cost > 0:
                 self.llm_cache_creation_cost_total.labels(
-                    org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+                    org_id=org_id,
+                    team_id=team_id,
+                    user_id=user_id,
+                    experiment_id=experiment_id,
+                    model=model,
                 ).inc(cache_creation_cost)
 
         except (ValueError, TypeError) as e:
             logger.debug(f"No cost data available for span: {e}")
 
         # Request count
         self.llm_requests_total.labels(
-            org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model, status=status
+            org_id=org_id,
+            team_id=team_id,
+            user_id=user_id,
+            experiment_id=experiment_id,
+            model=model,
+            status=status,
         ).inc()
 
         # Duration
         self.llm_request_duration_seconds.labels(
-            org_id=org_id, team_id=team_id, user_id=user_id, experiment_id=experiment_id, model=model
+            org_id=org_id,
+            team_id=team_id,
+            user_id=user_id,
+            experiment_id=experiment_id,
+            model=model,
         ).observe(duration)
 
     def _push_metrics(self):

charts/alphatrion/templates/ingress.yaml

@@ -28,14 +28,29 @@ spec:
     - host: {{ .host | quote }}
       http:
         paths:
-          {{- range .paths }}
-          - path: {{ .path }}
-            pathType: {{ .pathType }}
+          # API routes go to backend server
+          - path: /api
+            pathType: Prefix
+            backend:
+              service:
+                name: {{ include "alphatrion.server.fullname" $ }}
+                port:
+                  number: {{ $.Values.server.service.port }}
+          # GraphQL routes go to backend server
+          - path: /graphql
+            pathType: Prefix
+            backend:
+              service:
+                name: {{ include "alphatrion.server.fullname" $ }}
+                port:
+                  number: {{ $.Values.server.service.port }}
+          # Everything else (/, /static/*, SPA routes) goes to dashboard (static files)
+          - path: /
+            pathType: Prefix
             backend:
               service:
                 name: {{ include "alphatrion.dashboard.fullname" $ }}
                 port:
                   number: {{ $.Values.dashboard.service.port }}
-          {{- end }}
     {{- end }}
 {{- end }}

dashboard/src/components/layout/user-profile.tsx

@@ -7,14 +7,14 @@ import { useTeamContext } from '../../context/team-context';
 import { useOrganization } from '../../hooks/use-organization';
 import { cn } from '../../lib/utils';
 
-// Use runtime config (Kubernetes) > build-time env (Docker) > relative URL (local dev with proxy)
+// Use runtime config (Kubernetes) > build-time env (Docker) > localhost dev
 const getApiBaseUrl = () => {
   // @ts-ignore - window.ENV is injected at runtime by entrypoint.sh
   if (typeof window !== 'undefined' && window.ENV?.VITE_API_URL) {
     // @ts-ignore
     return window.ENV.VITE_API_URL;
   }
-  return import.meta.env.VITE_API_URL || '';
+  return import.meta.env.VITE_API_URL || 'http://localhost:8000';
 };
 
 export function UserProfile() {

dashboard/src/lib/graphql-client.ts

@@ -7,14 +7,14 @@ import axios from 'axios';
  * with queries and mutations for teams, experiments, runs, and metrics.
  */
 
-// Use runtime config (Kubernetes) > build-time env (Docker) > relative URL (local dev with proxy)
+// Use runtime config (Kubernetes) > build-time env (Docker) > localhost dev
 const getApiBaseUrl = () => {
   // @ts-ignore - window.ENV is injected at runtime by entrypoint.sh
   if (typeof window !== 'undefined' && window.ENV?.VITE_API_URL) {
     // @ts-ignore
     return window.ENV.VITE_API_URL;
   }
-  return import.meta.env.VITE_API_URL || '';
+  return import.meta.env.VITE_API_URL || 'http://localhost:8000';
 };
 
 const API_BASE_URL = getApiBaseUrl();

dashboard/src/pages/login.tsx

@@ -2,14 +2,14 @@ import { useState } from 'react';
 import { useNavigate } from 'react-router-dom';
 import logoImage from '../assets/logo.png';
 
-// Use runtime config (Kubernetes) > build-time env (Docker) > relative URL (local dev with proxy)
+// Use runtime config (Kubernetes) > build-time env (Docker) > localhost dev
 const getApiBaseUrl = () => {
   // @ts-ignore - window.ENV is injected at runtime by entrypoint.sh
   if (typeof window !== 'undefined' && window.ENV?.VITE_API_URL) {
     // @ts-ignore
     return window.ENV.VITE_API_URL;
   }
-  return import.meta.env.VITE_API_URL || '';
+  return import.meta.env.VITE_API_URL || 'http://localhost:8000';
 };
 
 export function LoginPage() {