Align Gunicorn configuration with Uvicorn

- centralize config in gunicorn_config.py
- fix log level handling in Gunicorn

gunicorn_config.py

@@ -2,17 +2,17 @@
 import os
 import logging
 from lightrag.kg.shared_storage import finalize_share_data
-from lightrag.api.utils_api import parse_args
 from lightrag.api.lightrag_server import LightragPathFilter
 
-# Parse command line arguments
-args = parse_args()
-
-# Determine worker count - from environment variable or command line arguments
-workers = int(os.getenv("WORKERS", args.workers))
+# 获取日志文件路径
+log_file_path = os.path.abspath(os.path.join(os.getcwd(), "lightrag.log"))
 
-# Binding address
-bind = f"{os.getenv('HOST', args.host)}:{os.getenv('PORT', args.port)}"
+# These variables will be set by run_with_gunicorn.py
+workers = None
+bind = None
+loglevel = None
+certfile = None
+keyfile = None
 
 # Enable preload_app option
 preload_app = True
@@ -24,18 +24,9 @@ worker_class = "uvicorn.workers.UvicornWorker"
 timeout = int(os.getenv("TIMEOUT", 120))
 keepalive = 5
 
-# Optional SSL configuration
-if args.ssl:
-    certfile = args.ssl_certfile
-    keyfile = args.ssl_keyfile
-
-# 获取日志文件路径
-log_file_path = os.path.abspath(os.path.join(os.getcwd(), "lightrag.log"))
-
 # Logging configuration
 errorlog = os.getenv("ERROR_LOG", log_file_path)  # 默认写入到 lightrag.log
 accesslog = os.getenv("ACCESS_LOG", log_file_path)  # 默认写入到 lightrag.log
-loglevel = os.getenv("LOG_LEVEL", "info")
 
 # 配置日志系统
 logconfig_dict = {
@@ -49,13 +40,11 @@ logconfig_dict = {
     'handlers': {
         'console': {
             'class': 'logging.StreamHandler',
-            'level': 'INFO',
             'formatter': 'standard',
             'stream': 'ext://sys.stdout'
         },
         'file': {
             'class': 'logging.handlers.RotatingFileHandler',
-            'level': 'INFO',
             'formatter': 'standard',
             'filename': log_file_path,
             'maxBytes': 10485760,  # 10MB
@@ -71,22 +60,22 @@ logconfig_dict = {
     'loggers': {
         'lightrag': {
             'handlers': ['console', 'file'],
-            'level': 'INFO',
+            'level': loglevel.upper() if loglevel else 'INFO',
             'propagate': False
         },
         'gunicorn': {
             'handlers': ['console', 'file'],
-            'level': 'INFO',
+            'level': loglevel.upper() if loglevel else 'INFO',
             'propagate': False
         },
         'gunicorn.error': {
             'handlers': ['console', 'file'],
-            'level': 'INFO',
+            'level': loglevel.upper() if loglevel else 'INFO',
             'propagate': False
         },
         'gunicorn.access': {
             'handlers': ['console', 'file'],
-            'level': 'INFO',
+            'level': loglevel.upper() if loglevel else 'INFO',
             'propagate': False,
             'filters': ['path_filter']
         }
@@ -143,6 +132,10 @@ def post_fork(server, worker):
     Executed after a worker has been forked.
     This is a good place to set up worker-specific configurations.
     """
+    # Set lightrag logger level in worker processes using gunicorn's loglevel
+    from lightrag.utils import logger
+    logger.setLevel(loglevel.upper())
+    
     # Disable uvicorn.error logger in worker processes
     uvicorn_error_logger = logging.getLogger("uvicorn.error")
     uvicorn_error_logger.setLevel(logging.CRITICAL)

lightrag/api/lightrag_server.py

@@ -86,7 +86,7 @@ class LightragPathFilter(logging.Filter):
 
 def create_app(args):
     # Setup logging
-    logger.setLevel(getattr(logging, args.log_level))
+    logger.setLevel(args.log_level)
     set_verbose_debug(args.verbose)
 
     # Verify that bindings are correctly setup
@@ -412,17 +412,10 @@ def create_app(args):
     return app
 
 
-def get_application():
+def get_application(args=None):
     """Factory function for creating the FastAPI application"""
-    # Get args from environment variable
-    args_json = os.environ.get("LIGHTRAG_ARGS")
-    if not args_json:
-        args = parse_args()  # Fallback to parsing args if env var not set
-    else:
-        import types
-
-        args = types.SimpleNamespace(**json.loads(args_json))
-
+    if args is None:
+        args = parse_args()
     return create_app(args)
 
 
@@ -513,10 +506,7 @@ def main():
     # Configure logging before parsing args
     configure_logging()
 
-    args = parse_args()
-    # Save args to environment variable for child processes
-    os.environ["LIGHTRAG_ARGS"] = json.dumps(vars(args))
-
+    args = parse_args(is_uvicorn_mode=True)
     display_splash_screen(args)
 
     # Create application instance directly instead of using factory function

lightrag/api/utils_api.py

@@ -111,10 +111,13 @@ def get_env_value(env_key: str, default: any, value_type: type = str) -> any:
         return default
 
 
-def parse_args() -> argparse.Namespace:
+def parse_args(is_uvicorn_mode: bool = False) -> argparse.Namespace:
     """
     Parse command line arguments with environment variable fallback
 
+    Args:
+        is_uvicorn_mode: Whether running under uvicorn mode
+
     Returns:
         argparse.Namespace: Parsed arguments
     """
@@ -287,9 +290,6 @@ def parse_args() -> argparse.Namespace:
 
     args = parser.parse_args()
 
-    # Check if running under uvicorn mode (not Gunicorn)
-    is_uvicorn_mode = "GUNICORN_CMD_ARGS" not in os.environ
-    
     # If in uvicorn mode and workers > 1, force it to 1 and log warning
     if is_uvicorn_mode and args.workers > 1:
         original_workers = args.workers

run_with_gunicorn.py

@@ -30,45 +30,10 @@ def main():
     # Register signal handlers for graceful shutdown
     signal.signal(signal.SIGINT, signal_handler)  # Ctrl+C
     signal.signal(signal.SIGTERM, signal_handler)  # kill command
-    # Create a parser to handle Gunicorn-specific parameters
-    parser = argparse.ArgumentParser(description="Start LightRAG server with Gunicorn")
-    parser.add_argument(
-        "--workers",
-        type=int,
-        help="Number of worker processes (overrides the default or config.ini setting)",
-    )
-    parser.add_argument(
-        "--timeout", type=int, help="Worker timeout in seconds (default: 120)"
-    )
-    parser.add_argument(
-        "--log-level",
-        choices=["debug", "info", "warning", "error", "critical"],
-        help="Gunicorn log level",
-    )
-
-    # Parse Gunicorn-specific arguments
-    gunicorn_args, remaining_args = parser.parse_known_args()
-
-    # Pass remaining arguments to LightRAG's parse_args
-    sys.argv = [sys.argv[0]] + remaining_args
-    args = parse_args()
-
-    # If workers specified, override args value
-    if gunicorn_args.workers:
-        args.workers = gunicorn_args.workers
-        os.environ["WORKERS"] = str(gunicorn_args.workers)
-
-    # If timeout specified, set environment variable
-    if gunicorn_args.timeout:
-        os.environ["TIMEOUT"] = str(gunicorn_args.timeout)
-
-    # If log-level specified, set environment variable
-    if gunicorn_args.log_level:
-        os.environ["LOG_LEVEL"] = gunicorn_args.log_level
-
-    # Save all LightRAG args to environment variable for worker processes
-    # This is the key step for passing arguments to lightrag_server.py
-    os.environ["LIGHTRAG_ARGS"] = json.dumps(vars(args))
+
+    # Parse all arguments using parse_args
+    args = parse_args(is_uvicorn_mode=False)
+
 
     # Display startup information
     display_splash_screen(args)
@@ -83,11 +48,6 @@ def main():
     print(f"Workers setting: {args.workers}")
     print("=" * 80 + "\n")
 
-    # Start application with Gunicorn using direct Python API
-    # Ensure WORKERS environment variable is set before importing gunicorn_config
-    if args.workers > 1:
-        os.environ["WORKERS"] = str(args.workers)
-
     # Import Gunicorn's StandaloneApplication
     from gunicorn.app.base import BaseApplication
 
@@ -136,51 +96,45 @@ def main():
                 "child_exit",
             }
 
-            # Import the gunicorn_config module directly
-            import importlib.util
+            # Import and configure the gunicorn_config module
+            import gunicorn_config
 
-            spec = importlib.util.spec_from_file_location(
-                "gunicorn_config", "gunicorn_config.py"
-            )
-            self.config_module = importlib.util.module_from_spec(spec)
-            spec.loader.exec_module(self.config_module)
+            # Set configuration variables in gunicorn_config
+            gunicorn_config.workers = int(os.getenv("WORKERS", args.workers))
+            gunicorn_config.bind = f"{os.getenv('HOST', args.host)}:{os.getenv('PORT', args.port)}"
+            gunicorn_config.loglevel = args.log_level.lower() if args.log_level else os.getenv("LOG_LEVEL", "info")
+            
+            # Set SSL configuration if enabled
+            if args.ssl:
+                gunicorn_config.certfile = args.ssl_certfile
+                gunicorn_config.keyfile = args.ssl_keyfile
 
-            # Set configuration options
-            for key in dir(self.config_module):
+            # Set configuration options from the module
+            for key in dir(gunicorn_config):
                 if key in valid_options:
-                    value = getattr(self.config_module, key)
-                    # Skip functions like on_starting
-                    if not callable(value):
+                    value = getattr(gunicorn_config, key)
+                    # Skip functions like on_starting and None values
+                    if not callable(value) and value is not None:
                         self.cfg.set(key, value)
                 # Set special hooks
                 elif key in special_hooks:
-                    value = getattr(self.config_module, key)
+                    value = getattr(gunicorn_config, key)
                     if callable(value):
                         self.cfg.set(key, value)
             
             # 确保正确加载 logconfig_dict
-            if hasattr(self.config_module, 'logconfig_dict'):
-                self.cfg.set('logconfig_dict', getattr(self.config_module, 'logconfig_dict'))
-
-            # Override with command line arguments if provided
-            if gunicorn_args.workers:
-                self.cfg.set("workers", gunicorn_args.workers)
-            if gunicorn_args.timeout:
-                self.cfg.set("timeout", gunicorn_args.timeout)
-            if gunicorn_args.log_level:
-                self.cfg.set("loglevel", gunicorn_args.log_level)
+            if hasattr(gunicorn_config, 'logconfig_dict'):
+                self.cfg.set('logconfig_dict', getattr(gunicorn_config, 'logconfig_dict'))
 
         def load(self):
             # Import the application
             from lightrag.api.lightrag_server import get_application
 
-            return get_application()
+            return get_application(args)
 
     # Create the application
     app = GunicornApp("")
 
-    # Directly call initialize_share_data with the correct workers value
-
     # Force workers to be an integer and greater than 1 for multi-process mode
     workers_count = int(args.workers)
     if workers_count > 1: