[Doc] Improve UX of `--enable-log-requests` (#35723)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

benchmarks/auto_tune/auto_tune.sh

@@ -85,7 +85,6 @@ start_server() {
     # Each argument and its value are separate elements.
     local common_args_array=(
         "$MODEL"
-        "--disable-log-requests"
         "--port" "8004"
         "--host" "$HOSTNAME"
         "--gpu-memory-utilization" "$gpu_memory_utilization"

benchmarks/multi_turn/README.md

@@ -7,7 +7,7 @@ First start serving your model
 ```bash
 export MODEL_PATH=/models/meta-llama/Meta-Llama-3.1-8B-Instruct/
 
-vllm serve $MODEL_PATH --served-model-name Llama --disable-log-requests
+vllm serve $MODEL_PATH --served-model-name Llama
 ```
 
 The variable `MODEL_PATH` should be a path to the model files (e.g. downloaded from huggingface).

vllm/engine/arg_utils.py

@@ -2187,14 +2187,10 @@ class AsyncEngineArgs(EngineArgs):
             "--enable-log-requests",
             action=argparse.BooleanOptionalAction,
             default=AsyncEngineArgs.enable_log_requests,
-            help="Enable logging requests.",
-        )
-        parser.add_argument(
-            "--disable-log-requests",
-            action=argparse.BooleanOptionalAction,
-            default=not AsyncEngineArgs.enable_log_requests,
-            help="[DEPRECATED] Disable logging requests.",
-            deprecated=True,
+            help="Enable logging request information, dependant on log level:\n"
+            "- INFO: Request ID, parameters and LoRA request.\n"
+            "- DEBUG: Prompt inputs (e.g: text, token IDs).\n"
+            "You can set the minimum log level via `VLLM_LOGGING_LEVEL`.",
         )
         current_platform.pre_register_and_update(parser)
         return parser

vllm/entrypoints/logger.py

@@ -18,6 +18,20 @@ class RequestLogger:
     def __init__(self, *, max_log_len: int | None) -> None:
         self.max_log_len = max_log_len
 
+        if not logger.isEnabledFor(logging.INFO):
+            logger.warning_once(
+                "`--enable-log-requests` is set but "
+                "the minimum log level is higher than INFO. "
+                "No request information will be logged."
+            )
+        elif not logger.isEnabledFor(logging.DEBUG):
+            logger.info_once(
+                "`--enable-log-requests` is set but "
+                "the minimum log level is higher than DEBUG. "
+                "Only limited information will be logged to minimize overhead. "
+                "To view more details, set `VLLM_LOGGING_LEVEL=DEBUG`."
+            )
+
     def log_inputs(
         self,
         request_id: str,

vllm/entrypoints/openai/cli_args.py

@@ -143,7 +143,8 @@ class BaseFrontendArgs:
     templates and other tokenizer configuration."""
     enable_log_outputs: bool = False
     """If set to True, log model outputs (generations).
-    Requires --enable-log-requests."""
+    Requires `--enable-log-requests`. As with `--enable-log-requests`,
+    information is only logged at INFO level at maximum."""
     enable_log_deltas: bool = True
     """If set to False, output deltas will not be logged. Relevant only if 
     --enable-log-outputs is set.