From fb9df50adad499d9259eecd16faad0e6e0677b98 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sat, 25 Jan 2025 22:14:40 +0800 Subject: [PATCH 01/13] Add conversation history support to chat API - Added HISTORY_TURNS env variable - Updated chat request data creation - Modified server to handle history - Added history to test cases --- .env.example | 1 + lightrag/api/lightrag_server.py | 26 ++++++++++++--- test_lightrag_ollama_chat.py | 59 ++++++++++++++++++++++++++++----- 3 files changed, 74 insertions(+), 12 deletions(-) diff --git a/.env.example b/.env.example index 944e7851..b652847b 100644 --- a/.env.example +++ b/.env.example @@ -43,6 +43,7 @@ MAX_ASYNC=4 MAX_TOKENS=32768 EMBEDDING_DIM=1024 MAX_EMBED_TOKENS=8192 +#HISTORY_TURNS=3 # Security (empty for no key) LIGHTRAG_API_KEY=your-secure-api-key-here diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index 533be818..b94e0d17 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -470,6 +470,13 @@ def parse_args() -> argparse.Namespace: help="Enable automatic scanning when the program starts", ) + parser.add_argument( + "--history-turns", + type=int, + default=get_env_value("HISTORY_TURNS", None, int), + help="Number of conversation history turns to include (default: from env or None)", + ) + args = parser.parse_args() return args @@ -1576,8 +1583,9 @@ def create_app(args): if not messages: raise HTTPException(status_code=400, detail="No messages provided") - # Get the last message as query + # Get the last message as query and previous messages as history query = messages[-1].content + conversation_history = messages[:-1] # 所有之前的消息作为历史记录 # Check for query prefix cleaned_query, mode = parse_query_mode(query) @@ -1585,9 +1593,19 @@ def create_app(args): start_time = time.time_ns() prompt_tokens = estimate_tokens(cleaned_query) - query_param = QueryParam( - mode=mode, stream=request.stream, only_need_context=False - ) + # 构建 query_param + param_dict = { + "mode": mode, + "stream": request.stream, + "only_need_context": False, + "conversation_history": conversation_history, + } + + # 如果设置了 history_turns，添加到参数中 + if args.history_turns is not None: + param_dict["history_turns"] = args.history_turns + + query_param = QueryParam(**param_dict) if request.stream: from fastapi.responses import StreamingResponse diff --git a/test_lightrag_ollama_chat.py b/test_lightrag_ollama_chat.py index d1e61d39..faf9485d 100644 --- a/test_lightrag_ollama_chat.py +++ b/test_lightrag_ollama_chat.py @@ -189,19 +189,32 @@ def get_base_url(endpoint: str = "chat") -> str: def create_chat_request_data( - content: str, stream: bool = False, model: str = None + content: str, + stream: bool = False, + model: str = None, + conversation_history: List[Dict[str, str]] = None, + history_turns: int = None, ) -> Dict[str, Any]: """Create chat request data Args: content: User message content stream: Whether to use streaming response model: Model name + conversation_history: List of previous conversation messages + history_turns: Number of history turns to include Returns: Dictionary containing complete chat request data """ + messages = conversation_history or [] + if history_turns is not None and conversation_history: + messages = messages[ + -2 * history_turns : + ] # Each turn has 2 messages (user + assistant) + messages.append({"role": "user", "content": content}) + return { "model": model or CONFIG["server"]["model"], - "messages": [{"role": "user", "content": content}], + "messages": messages, "stream": stream, } @@ -259,11 +272,25 @@ def run_test(func: Callable, name: str) -> None: def test_non_stream_chat() -> None: """Test non-streaming call to /api/chat endpoint""" url = get_base_url() - data = create_chat_request_data( - CONFIG["test_cases"]["basic"]["query"], stream=False - ) - # Send request + # Example conversation history + conversation_history = [ + {"role": "user", "content": "你好"}, + {"role": "assistant", "content": "你好!我是一个AI助手,很高兴为你服务。"}, + {"role": "user", "content": "西游记里有几个主要人物?"}, + { + "role": "assistant", + "content": "西游记的主要人物有唐僧、孙悟空、猪八戒、沙和尚这四位主角。", + }, + ] + + # Send request with conversation history and history turns + data = create_chat_request_data( + CONFIG["test_cases"]["basic"]["query"], + stream=False, + conversation_history=conversation_history, + history_turns=2, # Only include last 2 turns + ) response = make_request(url, data) # Print response @@ -297,9 +324,25 @@ def test_stream_chat() -> None: The last message will contain performance statistics, with done set to true. """ url = get_base_url() - data = create_chat_request_data(CONFIG["test_cases"]["basic"]["query"], stream=True) - # Send request and get streaming response + # Example conversation history + conversation_history = [ + {"role": "user", "content": "你好"}, + {"role": "assistant", "content": "你好!我是一个AI助手,很高兴为你服务。"}, + {"role": "user", "content": "西游记里有几个主要人物?"}, + { + "role": "assistant", + "content": "西游记的主要人物有唐僧、孙悟空、猪八戒、沙和尚这四位主角。", + }, + ] + + # Send request with conversation history and history turns + data = create_chat_request_data( + CONFIG["test_cases"]["basic"]["query"], + stream=True, + conversation_history=conversation_history, + history_turns=2, # Only include last 2 turns + ) response = make_request(url, data, stream=True) if OutputControl.is_verbose(): From 2719e07107db9f3f30b96d0776cdc8c828d78ece Mon Sep 17 00:00:00 2001 From: yangdx Date: Sat, 25 Jan 2025 22:33:09 +0800 Subject: [PATCH 02/13] Convert OllamaMessage to dict in conversation history --- lightrag/api/lightrag_server.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index b94e0d17..e000dc0e 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -1585,7 +1585,10 @@ def create_app(args): # Get the last message as query and previous messages as history query = messages[-1].content - conversation_history = messages[:-1] # 所有之前的消息作为历史记录 + # Convert OllamaMessage objects to dictionaries + conversation_history = [ + {"role": msg.role, "content": msg.content} for msg in messages[:-1] + ] # Check for query prefix cleaned_query, mode = parse_query_mode(query) From 86282ba4347cc8a0a698cd27958a71a8923f92a9 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sat, 25 Jan 2025 22:54:12 +0800 Subject: [PATCH 03/13] Fix history_context handling error in kg_query --- lightrag/operate.py | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/lightrag/operate.py b/lightrag/operate.py index af66eee6..0469fb7e 100644 --- a/lightrag/operate.py +++ b/lightrag/operate.py @@ -633,11 +633,8 @@ async def kg_query( # Process conversation history history_context = "" if query_param.conversation_history: - recent_history = query_param.conversation_history[ - -query_param.history_window_size : - ] - history_context = "\n".join( - [f"{turn['role']}: {turn['content']}" for turn in recent_history] + history_context = get_conversation_turns( + query_param.conversation_history, query_param.history_turns ) sys_prompt_temp = PROMPTS["rag_response"] From aa4bccb1bba947e892d018a716e6f8993f295a9f Mon Sep 17 00:00:00 2001 From: yangdx Date: Sat, 25 Jan 2025 23:38:45 +0800 Subject: [PATCH 04/13] Remove lightrag_api_open_webui_demo.py file --- examples/lightrag_api_open_webui_demo.py | 140 ----------------------- 1 file changed, 140 deletions(-) delete mode 100644 examples/lightrag_api_open_webui_demo.py diff --git a/examples/lightrag_api_open_webui_demo.py b/examples/lightrag_api_open_webui_demo.py deleted file mode 100644 index 88454da8..00000000 --- a/examples/lightrag_api_open_webui_demo.py +++ /dev/null @@ -1,140 +0,0 @@ -from datetime import datetime, timezone -from fastapi import FastAPI -from fastapi.responses import StreamingResponse -import inspect -import json -from pydantic import BaseModel -from typing import Optional - -import os -import logging -from lightrag import LightRAG, QueryParam -from lightrag.llm.ollama import ollama_model_complete, ollama_embed -from lightrag.utils import EmbeddingFunc - -import nest_asyncio - -WORKING_DIR = "./dickens" - -logging.basicConfig(format="%(levelname)s:%(message)s", level=logging.INFO) - -if not os.path.exists(WORKING_DIR): - os.mkdir(WORKING_DIR) - -rag = LightRAG( - working_dir=WORKING_DIR, - llm_model_func=ollama_model_complete, - llm_model_name="qwen2.5:latest", - llm_model_max_async=4, - llm_model_max_token_size=32768, - llm_model_kwargs={"host": "http://localhost:11434", "options": {"num_ctx": 32768}}, - embedding_func=EmbeddingFunc( - embedding_dim=1024, - max_token_size=8192, - func=lambda texts: ollama_embed( - texts=texts, embed_model="bge-m3:latest", host="http://127.0.0.1:11434" - ), - ), -) - -with open("./book.txt", "r", encoding="utf-8") as f: - rag.insert(f.read()) - -# Apply nest_asyncio to solve event loop issues -nest_asyncio.apply() - -app = FastAPI(title="LightRAG", description="LightRAG API open-webui") - - -# Data models -MODEL_NAME = "LightRAG:latest" - - -class Message(BaseModel): - role: Optional[str] = None - content: str - - -class OpenWebUIRequest(BaseModel): - stream: Optional[bool] = None - model: Optional[str] = None - messages: list[Message] - - -# API routes - - -@app.get("/") -async def index(): - return "Set Ollama link to http://ip:port/ollama in Open-WebUI Settings" - - -@app.get("/ollama/api/version") -async def ollama_version(): - return {"version": "0.4.7"} - - -@app.get("/ollama/api/tags") -async def ollama_tags(): - return { - "models": [ - { - "name": MODEL_NAME, - "model": MODEL_NAME, - "modified_at": "2024-11-12T20:22:37.561463923+08:00", - "size": 4683087332, - "digest": "845dbda0ea48ed749caafd9e6037047aa19acfcfd82e704d7ca97d631a0b697e", - "details": { - "parent_model": "", - "format": "gguf", - "family": "qwen2", - "families": ["qwen2"], - "parameter_size": "7.6B", - "quantization_level": "Q4_K_M", - }, - } - ] - } - - -@app.post("/ollama/api/chat") -async def ollama_chat(request: OpenWebUIRequest): - resp = rag.query( - request.messages[-1].content, param=QueryParam(mode="hybrid", stream=True) - ) - if inspect.isasyncgen(resp): - - async def ollama_resp(chunks): - async for chunk in chunks: - yield ( - json.dumps( - { - "model": MODEL_NAME, - "created_at": datetime.now(timezone.utc).strftime( - "%Y-%m-%dT%H:%M:%S.%fZ" - ), - "message": { - "role": "assistant", - "content": chunk, - }, - "done": False, - }, - ensure_ascii=False, - ).encode("utf-8") - + b"\n" - ) # the b"\n" is important - - return StreamingResponse(ollama_resp(resp), media_type="application/json") - else: - return resp - - -@app.get("/health") -async def health_check(): - return {"status": "healthy"} - - -if __name__ == "__main__": - import uvicorn - - uvicorn.run(app, host="0.0.0.0", port=8020) From 22742ec9cb8f0c3dc991056f2892505d6850bfda Mon Sep 17 00:00:00 2001 From: yangdx Date: Sat, 25 Jan 2025 23:46:29 +0800 Subject: [PATCH 05/13] fix: remove outdated Ollama model config notes - Remove legacy configuration instructions for Open WebUI tasks - Ollama API can properly bypass conversation metadata generation --- lightrag/api/README.md | 2 -- 1 file changed, 2 deletions(-) diff --git a/lightrag/api/README.md b/lightrag/api/README.md index 89906006..4e818242 100644 --- a/lightrag/api/README.md +++ b/lightrag/api/README.md @@ -94,8 +94,6 @@ For example, chat message "/mix 唐僧有几个徒弟" will trigger a mix mode q After starting the lightrag-server, you can add an Ollama-type connection in the Open WebUI admin pannel. And then a model named lightrag:latest will appear in Open WebUI's model management interface. Users can then send queries to LightRAG through the chat interface. -To prevent Open WebUI from using LightRAG when generating conversation titles, go to Admin Panel > Interface > Set Task Model and change both Local Models and External Models to any option except "Current Model". - ## Configuration LightRAG can be configured using either command-line arguments or environment variables. When both are provided, command-line arguments take precedence over environment variables. From e4e42a8ec43811a63fc445c803bb28b0b11758f9 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 02:31:16 +0800 Subject: [PATCH 06/13] Fetch chunk size and chunk overlap size from .env file --- .env.example | 2 ++ lightrag/api/lightrag_server.py | 8 ++++---- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/.env.example b/.env.example index b652847b..82b9ca70 100644 --- a/.env.example +++ b/.env.example @@ -44,6 +44,8 @@ MAX_TOKENS=32768 EMBEDDING_DIM=1024 MAX_EMBED_TOKENS=8192 #HISTORY_TURNS=3 +#CHUNK_SIZE=1200 +#CHUNK_OVERLAP_SIZE=100 # Security (empty for no key) LIGHTRAG_API_KEY=your-secure-api-key-here diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index e000dc0e..da15d212 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -383,14 +383,14 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--chunk_size", - default=1200, - help="chunk token size default 1200", + default=get_env_value("CHUNK_SIZE", 1200), + help="chunk chunk size default 1200", ) parser.add_argument( "--chunk_overlap_size", - default=100, - help="chunk token size default 1200", + default=get_env_value("CHUNK_OVERLAP_SIZE", 100), + help="chunk overlap size default 100", ) def timeout_type(value): From 9f80c1904ff11e6a0ba68779bdfdc42285beb4be Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 05:09:42 +0800 Subject: [PATCH 07/13] Refactoring command line argurements handling logic, add more RAG config to splash screen --- lightrag/api/lightrag_server.py | 42 ++++++++++++++++++++------------- 1 file changed, 26 insertions(+), 16 deletions(-) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index da15d212..2d183b73 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -200,8 +200,14 @@ def display_splash_screen(args: argparse.Namespace) -> None: ASCIIColors.yellow(f"{args.max_async}") ASCIIColors.white(" ├─ Max Tokens: ", end="") ASCIIColors.yellow(f"{args.max_tokens}") - ASCIIColors.white(" └─ Max Embed Tokens: ", end="") + ASCIIColors.white(" ├─ Max Embed Tokens: ", end="") ASCIIColors.yellow(f"{args.max_embed_tokens}") + ASCIIColors.white(" ├─ Chunk Size: ", end="") + ASCIIColors.yellow(f"{args.chunk_size}") + ASCIIColors.white(" ├─ Chunk Overlap Size: ", end="") + ASCIIColors.yellow(f"{args.chunk_overlap_size}") + ASCIIColors.white(" └─ History Turns: ", end="") + ASCIIColors.yellow(f"{args.history_turns}") # System Configuration ASCIIColors.magenta("\n🛠️ System Configuration:") @@ -294,7 +300,7 @@ def parse_args() -> argparse.Namespace: description="LightRAG FastAPI Server with separate working and input directories" ) - # Bindings (with env var support) + # Bindings configuration parser.add_argument( "--llm-binding", default=get_env_value("LLM_BINDING", "ollama"), @@ -306,8 +312,6 @@ def parse_args() -> argparse.Namespace: help="Embedding binding to be used. Supported: lollms, ollama, openai (default: from env or ollama)", ) - # Parse temporary args for host defaults - temp_args, _ = parser.parse_known_args() # Server configuration parser.add_argument( @@ -335,13 +339,13 @@ def parse_args() -> argparse.Namespace: ) # LLM Model configuration - default_llm_host = get_env_value( - "LLM_BINDING_HOST", get_default_host(temp_args.llm_binding) - ) parser.add_argument( "--llm-binding-host", - default=default_llm_host, - help=f"llm server host URL (default: from env or {default_llm_host})", + default=get_env_value("LLM_BINDING_HOST", None), + help="LLM server host URL. If not provided, defaults based on llm-binding:\n" + + "- ollama: http://localhost:11434\n" + + "- lollms: http://localhost:9600\n" + + "- openai: https://api.openai.com/v1", ) default_llm_api_key = get_env_value("LLM_BINDING_API_KEY", None) @@ -359,13 +363,13 @@ def parse_args() -> argparse.Namespace: ) # Embedding model configuration - default_embedding_host = get_env_value( - "EMBEDDING_BINDING_HOST", get_default_host(temp_args.embedding_binding) - ) parser.add_argument( "--embedding-binding-host", - default=default_embedding_host, - help=f"embedding server host URL (default: from env or {default_embedding_host})", + default=get_env_value("EMBEDDING_BINDING_HOST", None), + help="Embedding server host URL. If not provided, defaults based on embedding-binding:\n" + + "- ollama: http://localhost:11434\n" + + "- lollms: http://localhost:9600\n" + + "- openai: https://api.openai.com/v1", ) default_embedding_api_key = get_env_value("EMBEDDING_BINDING_API_KEY", "") @@ -641,8 +645,7 @@ def get_api_key_dependency(api_key: Optional[str]): def create_app(args): - # Verify that bindings arer correctly setup - + # Verify that bindings are correctly setup if args.llm_binding not in [ "lollms", "ollama", @@ -655,6 +658,13 @@ def create_app(args): if args.embedding_binding not in ["lollms", "ollama", "openai", "azure_openai"]: raise Exception("embedding binding not supported") + # Set default hosts if not provided + if args.llm_binding_host is None: + args.llm_binding_host = get_default_host(args.llm_binding) + + if args.embedding_binding_host is None: + args.embedding_binding_host = get_default_host(args.embedding_binding) + # Add SSL validation if args.ssl: if not args.ssl_certfile or not args.ssl_keyfile: From 25a58a6545d183e305fa07c771083252fce47135 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 05:10:57 +0800 Subject: [PATCH 08/13] Fix linting --- lightrag/api/lightrag_server.py | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index 2d183b73..2c7e0a54 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -312,7 +312,6 @@ def parse_args() -> argparse.Namespace: help="Embedding binding to be used. Supported: lollms, ollama, openai (default: from env or ollama)", ) - # Server configuration parser.add_argument( "--host", @@ -342,10 +341,10 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--llm-binding-host", default=get_env_value("LLM_BINDING_HOST", None), - help="LLM server host URL. If not provided, defaults based on llm-binding:\n" + - "- ollama: http://localhost:11434\n" + - "- lollms: http://localhost:9600\n" + - "- openai: https://api.openai.com/v1", + help="LLM server host URL. If not provided, defaults based on llm-binding:\n" + + "- ollama: http://localhost:11434\n" + + "- lollms: http://localhost:9600\n" + + "- openai: https://api.openai.com/v1", ) default_llm_api_key = get_env_value("LLM_BINDING_API_KEY", None) @@ -366,10 +365,10 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--embedding-binding-host", default=get_env_value("EMBEDDING_BINDING_HOST", None), - help="Embedding server host URL. If not provided, defaults based on embedding-binding:\n" + - "- ollama: http://localhost:11434\n" + - "- lollms: http://localhost:9600\n" + - "- openai: https://api.openai.com/v1", + help="Embedding server host URL. If not provided, defaults based on embedding-binding:\n" + + "- ollama: http://localhost:11434\n" + + "- lollms: http://localhost:9600\n" + + "- openai: https://api.openai.com/v1", ) default_embedding_api_key = get_env_value("EMBEDDING_BINDING_API_KEY", "") @@ -661,7 +660,7 @@ def create_app(args): # Set default hosts if not provided if args.llm_binding_host is None: args.llm_binding_host = get_default_host(args.llm_binding) - + if args.embedding_binding_host is None: args.embedding_binding_host = get_default_host(args.embedding_binding) From f8d26cb19303dcbd2299244a377c72794a5feead Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 05:19:51 +0800 Subject: [PATCH 09/13] Update default history turns to 3 --- lightrag/api/lightrag_server.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index 2c7e0a54..de6cb794 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -476,8 +476,8 @@ def parse_args() -> argparse.Namespace: parser.add_argument( "--history-turns", type=int, - default=get_env_value("HISTORY_TURNS", None, int), - help="Number of conversation history turns to include (default: from env or None)", + default=get_env_value("HISTORY_TURNS", 3, int), + help="Number of conversation history turns to include (default: from env or 3)", ) args = parser.parse_args() From f045fc3d590ad70d1f78697cac22cc5d54aec7e5 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 11:36:24 +0800 Subject: [PATCH 10/13] Update API endpoint documentation --- lightrag/api/lightrag_server.py | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index de6cb794..bbf72374 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -1446,7 +1446,10 @@ def create_app(args): @app.post("/api/generate") async def generate(raw_request: Request, request: OllamaGenerateRequest): - """Handle generate completion requests""" + """Handle generate completion requests + For compatiblity purpuse, the request is not processed by LightRAG, + and will be handled by underlying LLM model. + """ try: query = request.prompt start_time = time.time_ns() @@ -1585,7 +1588,10 @@ def create_app(args): @app.post("/api/chat") async def chat(raw_request: Request, request: OllamaChatRequest): - """Handle chat completion requests""" + """Process chat completion requests. + Routes user queries through LightRAG by selecting query mode based on prefix indicators. + Detects and forwards OpenWebUI session-related requests (for meta data generation task) directly to LLM. + """ try: # Get all messages messages = request.messages @@ -1605,7 +1611,6 @@ def create_app(args): start_time = time.time_ns() prompt_tokens = estimate_tokens(cleaned_query) - # 构建 query_param param_dict = { "mode": mode, "stream": request.stream, @@ -1613,7 +1618,6 @@ def create_app(args): "conversation_history": conversation_history, } - # 如果设置了 history_turns，添加到参数中 if args.history_turns is not None: param_dict["history_turns"] = args.history_turns From c1edef7119fdaad515d2dc298120d789764440f9 Mon Sep 17 00:00:00 2001 From: yangdx Date: Sun, 26 Jan 2025 23:21:32 +0800 Subject: [PATCH 11/13] Increase timeout for LightRAG Ollama chat --- test_lightrag_ollama_chat.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/test_lightrag_ollama_chat.py b/test_lightrag_ollama_chat.py index faf9485d..6982d44b 100644 --- a/test_lightrag_ollama_chat.py +++ b/test_lightrag_ollama_chat.py @@ -104,7 +104,7 @@ DEFAULT_CONFIG = { "host": "localhost", "port": 9621, "model": "lightrag:latest", - "timeout": 30, + "timeout": 120, "max_retries": 3, "retry_delay": 1, }, From 01288debd148f39eca4545b8b1a291f5930b028a Mon Sep 17 00:00:00 2001 From: yangdx Date: Mon, 27 Jan 2025 02:45:44 +0800 Subject: [PATCH 12/13] Ensure splash output flush to system log --- lightrag/api/lightrag_server.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index bbf72374..d700d5c5 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -17,6 +17,7 @@ import shutil import aiofiles from ascii_colors import trace_exception, ASCIIColors import os +import sys import configparser from fastapi import Depends, Security @@ -287,6 +288,8 @@ def display_splash_screen(args: argparse.Namespace) -> None: ASCIIColors.green("Server is ready to accept connections! 🚀\n") + # Ensure splash output flush to system log + sys.stdout.flush() def parse_args() -> argparse.Namespace: """ From 03604d3186c9ba93fbf203fb514b7c0b978a2adb Mon Sep 17 00:00:00 2001 From: yangdx Date: Mon, 27 Jan 2025 02:46:21 +0800 Subject: [PATCH 13/13] Fix linting --- lightrag/api/lightrag_server.py | 1 + 1 file changed, 1 insertion(+) diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py index d700d5c5..3f245e9f 100644 --- a/lightrag/api/lightrag_server.py +++ b/lightrag/api/lightrag_server.py @@ -291,6 +291,7 @@ def display_splash_screen(args: argparse.Namespace) -> None: # Ensure splash output flush to system log sys.stdout.flush() + def parse_args() -> argparse.Namespace: """ Parse command line arguments with environment variable fallback