feat(cache): enhance cache monitoring endpoints and handler integrations

src/api/handlers/base/base_handler.py

@@ -395,3 +395,24 @@ class BaseMessageHandler:
 
         # 创建后台任务，不阻塞当前流
         asyncio.create_task(_do_update())
+
+    def _log_request_error(self, message: str, error: Exception) -> None:
+        """记录请求错误日志，对业务异常不打印堆栈
+
+        Args:
+            message: 错误消息前缀
+            error: 异常对象
+        """
+        from src.core.exceptions import (
+            ProviderException,
+            QuotaExceededException,
+            RateLimitException,
+            ModelNotSupportedException,
+        )
+
+        if isinstance(error, (ProviderException, QuotaExceededException, RateLimitException, ModelNotSupportedException)):
+            # 业务异常：简洁日志，不打印堆栈
+            logger.error(f"{message}: [{type(error).__name__}] {error}")
+        else:
+            # 未知异常：完整堆栈
+            logger.exception(f"{message}: {error}")

src/api/handlers/base/chat_handler_base.py

@@ -382,7 +382,7 @@ class ChatHandlerBase(BaseMessageHandler, ABC):
             )
 
         except Exception as e:
-            logger.exception(f"流式请求失败: {e}")
+            self._log_request_error("流式请求失败", e)
             await self._record_stream_failure(ctx, e, original_headers, original_request_body)
             raise
 

src/api/handlers/base/cli_handler_base.py

@@ -413,20 +413,7 @@ class CliMessageHandlerBase(BaseMessageHandler):
             )
 
         except Exception as e:
-            # 对于已知的业务异常，只记录简洁的错误信息，不输出完整堆栈
-            from src.core.exceptions import (
-                ProviderException,
-                QuotaExceededException,
-                RateLimitException,
-                ModelNotSupportedException,
-            )
-
-            if isinstance(e, (ProviderException, QuotaExceededException, RateLimitException, ModelNotSupportedException)):
-                # 业务异常：简洁日志
-                logger.error(f"流式请求失败: [{type(e).__name__}] {e}")
-            else:
-                # 未知异常：完整堆栈
-                logger.exception(f"流式请求失败: {e}")
+            self._log_request_error("流式请求失败", e)
             await self._record_stream_failure(ctx, e, original_headers, original_request_body)
             raise