fix(knowledge): 全量修复多 worker 元数据竞态问题

1. collection.py: create/delete/update/exists 方法开头加 _load_metadata()
2. list_collections: 去掉磁盘扫描自动补充和 stale 清理逻辑,
   改为以元数据为唯一真相源,异常集合只跳过不删不改
3. manager.py: _load/_save_metadata 加 fcntl 文件锁防并发写覆盖,
   Windows 兼容(fcntl try/import 包裹)
This commit is contained in:
lacerate551
2026-06-10 11:17:57 +08:00
parent 431af0217a
commit 148559ee3c
2 changed files with 46 additions and 47 deletions

View File

@@ -29,6 +29,11 @@
import os
import json
import threading
try:
import fcntl
_HAS_FCNTL = True
except ImportError:
_HAS_FCNTL = False
from typing import List, Dict, Optional, Tuple
from pathlib import Path
import logging
@@ -134,22 +139,36 @@ class KnowledgeBaseManager(
logger.info(f"知识库管理器初始化完成,路径: {self.base_path},发现 {len(existing_kbs)} 个向量库: {existing_kbs}")
def _load_metadata(self) -> dict:
"""加载元数据"""
"""加载元数据(带文件锁)"""
metadata_path = os.path.join(self.base_path, KB_METADATA_FILE)
if os.path.exists(metadata_path):
try:
with open(metadata_path, 'r', encoding='utf-8') as f:
return json.load(f)
if _HAS_FCNTL:
fcntl.flock(f.fileno(), fcntl.LOCK_SH)
try:
return json.load(f)
finally:
if _HAS_FCNTL:
fcntl.flock(f.fileno(), fcntl.LOCK_UN)
except Exception as e:
logger.error(f"加载元数据失败: {e}")
return {"collections": {}}
def _save_metadata(self):
"""保存元数据"""
"""保存元数据(带文件锁,防止并发写入覆盖)"""
metadata_path = os.path.join(self.base_path, KB_METADATA_FILE)
try:
with open(metadata_path, 'w', encoding='utf-8') as f:
json.dump(self._metadata, f, ensure_ascii=False, indent=2)
if _HAS_FCNTL:
fcntl.flock(f.fileno(), fcntl.LOCK_EX)
try:
json.dump(self._metadata, f, ensure_ascii=False, indent=2)
f.flush()
os.fsync(f.fileno())
finally:
if _HAS_FCNTL:
fcntl.flock(f.fileno(), fcntl.LOCK_UN)
except Exception as e:
logger.error(f"保存元数据失败: {e}")