[misc][frontend] log all available endpoints (vllm-project#6195)

Co-authored-by: Cody Yu <[email protected]>

vllm/entrypoints/api_server.py

@@ -16,10 +16,13 @@
 
 from vllm.engine.arg_utils import AsyncEngineArgs
 from vllm.engine.async_llm_engine import AsyncLLMEngine
+from vllm.logger import init_logger
 from vllm.sampling_params import SamplingParams
 from vllm.usage.usage_lib import UsageContext
 from vllm.utils import FlexibleArgumentParser, random_uuid
 
+logger = init_logger("vllm.entrypoints.api_server")
+
 TIMEOUT_KEEP_ALIVE = 5  # seconds.
 app = FastAPI()
 engine = None
@@ -107,6 +110,14 @@ async def stream_results() -> AsyncGenerator[bytes, None]:
         engine_args, usage_context=UsageContext.API_SERVER)
 
     app.root_path = args.root_path
+
+    logger.info("Available routes are:")
+    for route in app.routes:
+        if not hasattr(route, 'methods'):
+            continue
+        methods = ', '.join(route.methods)
+        logger.info("Route: %s, Methods: %s", route.path, methods)
+
     uvicorn.run(app,
                 host=args.host,
                 port=args.port,

vllm/entrypoints/openai/api_server.py

@@ -240,6 +240,14 @@ async def authentication(request: Request, call_next):
     openai_serving_embedding = OpenAIServingEmbedding(engine, model_config,
                                                       served_model_names)
     app.root_path = args.root_path
+
+    logger.info("Available routes are:")
+    for route in app.routes:
+        if not hasattr(route, 'methods'):
+            continue
+        methods = ', '.join(route.methods)
+        logger.info("Route: %s, Methods: %s", route.path, methods)
+
     uvicorn.run(app,
                 host=args.host,
                 port=args.port,