Update app.py

app.py

@@ -30,6 +30,15 @@ t5_model = None
 mistral = None
 models_loaded = False
 
+# Root endpoint to confirm server is running
+@app.get("/")
+async def root():
+    logger.info("Root endpoint called at %s", time.time())
+    return JSONResponse(
+        content={"message": "MGZON Smart Assistant is running"},
+        headers={"Cache-Control": "no-cache", "Connection": "keep-alive"}
+    )
+
 # Health check endpoint
 @app.get("/health")
 async def health_check():
@@ -80,7 +89,7 @@ async def load_models():
         mistral = Llama(
             model_path=gguf_path,
             n_ctx=2048,
-            n_threads=4,
+            n_threads=2,  # قللنا الـ threads إلى 2 عشان نقلل الحمل أكتر
             n_batch=512,
             verbose=True
         )
@@ -96,7 +105,7 @@ async def load_models():
 # Run model loading in the background
 @app.on_event("startup")
 async def startup_event():
-    logger.info("Startup event triggered")
+    logger.info("Startup event triggered at %s", time.time())
     asyncio.create_task(load_models())
 
 # Define request schema
@@ -107,7 +116,7 @@ class AskRequest(BaseModel):
 # Endpoint: /ask
 @app.post("/ask")
 async def ask(req: AskRequest):
-    logger.info(f"Received ask request: {req.question}")
+    logger.info(f"Received ask request: {req.question} at {time.time()}")
     if not models_loaded:
         logger.error("Models not loaded yet")
         raise HTTPException(status_code=503, detail="Models are still loading, please try again later")
@@ -139,4 +148,11 @@ async def ask(req: AskRequest):
 
 # Run the app
 if __name__ == "__main__":
-    uvicorn.run(app, host="0.0.0.0", port=8080, log_level="info")
+    uvicorn.run(
+        app,
+        host="0.0.0.0",
+        port=8080,
+        log_level="info",
+        workers=1,  # استخدام worker واحد عشان نقلل الحمل
+        timeout_keep_alive=30  # ضمان استجابة سريعة
+    )