Add deployment tuning and chat load test

ai_knowledge_base_v2/apps/backend/app/api/chat.py

@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import asyncio
 import json
+import logging
 from collections.abc import AsyncIterator
 
 from fastapi import APIRouter, Depends, HTTPException, Query
@@ -31,6 +32,7 @@ from app.services.chat_queue_service import load_chat_queue_config
 from app.services.chat_stream_service import ChatStreamService
 
 router = APIRouter()
+logger = logging.getLogger(__name__)
 
 
 @router.post("/session")
@@ -155,6 +157,7 @@ async def _chat_stream(payload: ChatCompletionRequest, db: Session, current_user
     except asyncio.CancelledError:
         raise
     except Exception:
+        logger.exception("Chat stream failed")
         yield _sse_event("error", message="AI 回复生成失败，请稍后再试。")
     finally:
         if acquired:

ai_knowledge_base_v2/apps/backend/app/core/config.py

@@ -26,6 +26,10 @@ class Settings(BaseSettings):
 
     database_url: str = "mysql+pymysql://ai_kb:ai_kb@127.0.0.1:3306/ai_knowledge_base_v2?charset=utf8mb4"
     redis_url: str = ""
+    db_pool_size: int = 10
+    db_max_overflow: int = 20
+    db_pool_timeout: int = 30
+    db_pool_recycle: int = 1800
     auto_create_tables: bool = False
 
     jwt_secret_key: str = "change-me-in-production"

ai_knowledge_base_v2/apps/backend/app/core/database.py

@@ -10,7 +10,15 @@ from app.models import Base
 
 settings = get_settings()
 
-engine = create_engine(settings.database_url, pool_pre_ping=True, future=True)
+engine = create_engine(
+    settings.database_url,
+    pool_pre_ping=True,
+    pool_size=settings.db_pool_size,
+    max_overflow=settings.db_max_overflow,
+    pool_timeout=settings.db_pool_timeout,
+    pool_recycle=settings.db_pool_recycle,
+    future=True,
+)
 SessionLocal = sessionmaker(bind=engine, autoflush=False, autocommit=False, future=True)