[Bug][Fixed] mcp服务异常mcp.shared.exceptions.McpError: Session terminated

Start with astrbot/core/agent/mcp_client.py, especially call_tool_with_reconnect and the retry and exception-handling points identified in the report. Reproduce with the listed streamable_http configuration and review the Session terminated logs. Done means the MCP service reconnects and the tool call succeeds after the container is restarted.

问题描述

docker使用镜像源部署，添加https://mcpmarket.cn的mcp服务的时候发现mcp.shared.exceptions.McpError: Session terminated。
临时修复：容器内运行

cd /tmp && cat > fix_mcp.py << 'PYEOF'
import io

path = "/AstrBot/astrbot/core/agent/mcp_client.py"
src = io.open(path, encoding="utf-8").read()

# 1. 顶部加 import
if "from mcp.shared.exceptions import McpError" not in src:
    assert "import httpx\n" in src, "import anchor not found"
    src = src.replace(
        "import httpx\n",
        "import httpx\nfrom mcp.shared.exceptions import McpError\n",
        1,
    )

# 2. 扩大重试条件
old_retry = "retry=retry_if_exception_type(anyio.ClosedResourceError),"
new_retry = "retry=retry_if_exception_type((anyio.ClosedResourceError, TimeoutError, McpError)),"
assert src.count(old_retry) == 1, "retry anchor not found"
src = src.replace(old_retry, new_retry)

# 3. 扩大异常捕获并加重连保护
old_exc = """            except anyio.ClosedResourceError:
                logger.warning(
                    f"MCP tool {tool_name} call failed (ClosedResourceError), attempting to reconnect..."
                )
                # Attempt to reconnect
                await self._reconnect()
                # Reraise the exception to trigger tenacity retry
                raise"""
new_exc = """            except (anyio.ClosedResourceError, TimeoutError, McpError) as e:
                logger.warning(
                    f"MCP tool {tool_name} call failed ({type(e).__name__}: {e}), attempting to reconnect..."
                )
                # Attempt to reconnect (best effort, don't mask the original error)
                try:
                    await self._reconnect()
                except Exception as re:
                    logger.error(f"Reconnect failed: {re}")
                # Reraise the exception to trigger tenacity retry
                raise"""
assert src.count(old_exc) == 1, "except anchor not found"
src = src.replace(old_exc, new_exc)

io.open(path, "w", encoding="utf-8").write(src)
print("OK: 三处修改全部完成")
PYEOF
python fix_mcp.py

重启容器后得到解决

如何复现？

正常docker部署，用镜像源m.daocloud.io/docker.io/soulter/astrbot:latest+m.daocloud.io/docker.io/mlikiowa/napcat-docker:latest（版本对应为4.27.4），添加配置：
root@663fab43182d:/AstrBot/data# cat mcp_server.json
{
"mcpServers": {
"高德地图": {
"active": true,
"transport": "streamable_http",
"url": "https://mcpmarket.cn/mcp/688537b7d00f731bbe02e03e",
"headers": {},
"timeout": 5,
"sse_read_timeout": 300
},
"天气查询": {
"active": true,
"transport": "streamable_http",
"url": "https://mcpmarket.cn/mcp/5b3f77de1dac10ca4f373352",
"headers": {},
"timeout": 5,
"sse_read_timeout": 300
},
"12306服务": {
"active": true,
"transport": "streamable_http",
"url": "https://mcpmarket.cn/mcp/6e628a618abaac31b04789e4",
"headers": {},
"timeout": 5,
"sse_read_timeout": 300
},
"MarkitDown": {
"active": true,
"transport": "streamable_http",
"url": "https://mcpmarket.cn/mcp/bfd0296c9522e068abfc6178",
"headers": {},
"timeout": 5,
"sse_read_timeout": 300
},
"百度搜索": {
"active": true,
"transport": "streamable_http",
"url": "https://mcpmarket.cn/mcp/559f4b2ecdd8d3b2efd477fa",
"headers": {},
"timeout": 5,
"sse_read_timeout": 300
}
}
}
在对话中服务不可用。

AstrBot 版本

v4.27.4

操作系统

Linux

部署方式

docker

使用的消息平台适配器

Onebot v11

错误日志

[2026-08-27 07:00:07.922] [Core]
[INFO]
[runners.tool_loop_agent_runner:1152]: 使用工具：get_weather_forecast，参数：{'latitude': 31.4914, 'longitude': 120.3119, 'days': 1, 'hourly_vars': 'temperature_2m,precipitation,windspeed_10m,weathercode,relative_humidity_2m'}
[2026-08-27 07:00:08.078] [Core]
[WARN]
[v4.27.4] [runners.tool_loop_agent_runner:1324]: Traceback (most recent call last):
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1206, in _handle_function_tools
    async for resp in self._iter_tool_executor_results(executor):  # type: ignore
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1561, in _iter_tool_executor_results
    yield next_result_task.result()
          ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1531, in _next_executor_result
    return await anext(executor)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 185, in execute
    async for r in cls._execute_local(tool, run_context, **tool_args):
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 666, in _execute_local
    resp = await asyncio.wait_for(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/tasks.py", line 520, in wait_for
    return await fut
           ^^^^^^^^^
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 807, in call_local_llm_tool
    ret = await ready_to_call
          ^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/provider/func_tool_manager.py", line 269, in call
    return await self._wrapped.call(context, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 817, in call
    return await self.mcp_client.call_tool_with_reconnect(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 774, in call_tool_with_reconnect
    return await _call_with_retry()
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 193, in async_wrapped
    return await copy(fn, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 112, in __call__
    do = await self.iter(retry_state=retry_state)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 157, in iter
    result = await action(retry_state)
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/_utils.py", line 111, in inner
    return call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/__init__.py", line 393, in <lambda>
    self._add_action_func(lambda rs: rs.outcome.result())
                                     ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/_base.py", line 449, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 116, in __call__
    result = await fn(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 760, in _call_with_retry
    return await self.session.call_tool(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/mcp/client/session.py", line 401, in call_tool
    result = await self.send_request(
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/mcp/shared/session.py", line 306, in send_request
    raise McpError(response_or_error.error)
mcp.shared.exceptions.McpError: Session terminated
[2026-08-27 07:00:08.079] [Core]
[INFO]
[runners.tool_loop_agent_runner:1349]: Tool `get_weather_forecast` Result: error: Session terminated
[2026-08-28 07:00:05.933] [Core]
[INFO]
[runners.tool_loop_agent_runner:1096]: Agent 使用工具: ['get-current-date', 'get_weather_forecast']
[2026-08-28 07:00:05.933] [Core]
[INFO]
[runners.tool_loop_agent_runner:1152]: 使用工具：get-current-date，参数：{}
[2026-08-28 07:02:05.975] [Core]
[WARN]
[v4.27.4] [runners.tool_loop_agent_runner:1324]: Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/anyio/streams/memory.py", line 118, in receive
    return self.receive_nowait()
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/anyio/streams/memory.py", line 113, in receive_nowait
    raise WouldBlock
anyio.WouldBlock

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/usr/local/lib/python3.12/asyncio/tasks.py", line 520, in wait_for
    return await fut
           ^^^^^^^^^
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 807, in call_local_llm_tool
    ret = await ready_to_call
          ^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/provider/func_tool_manager.py", line 269, in call
    return await self._wrapped.call(context, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 817, in call
    return await self.mcp_client.call_tool_with_reconnect(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 774, in call_tool_with_reconnect
    return await _call_with_retry()
           ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 193, in async_wrapped
    return await copy(fn, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 112, in __call__
    do = await self.iter(retry_state=retry_state)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 157, in iter
    result = await action(retry_state)
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/_utils.py", line 111, in inner
    return call(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/tenacity/__init__.py", line 393, in <lambda>
    self._add_action_func(lambda rs: rs.outcome.result())
                                     ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/_base.py", line 449, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/usr/local/lib/python3.12/site-packages/tenacity/asyncio/__init__.py", line 116, in __call__
    result = await fn(*args, **kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/mcp_client.py", line 760, in _call_with_retry
    return await self.session.call_tool(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/mcp/client/session.py", line 401, in call_tool
    result = await self.send_request(
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/mcp/shared/session.py", line 292, in send_request
    response_or_error = await response_stream_reader.receive()
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/anyio/streams/memory.py", line 126, in receive
    await receive_event.wait()
  File "/usr/local/lib/python3.12/site-packages/anyio/_backends/_asyncio.py", line 1866, in wait
    await self._event.wait()
  File "/usr/local/lib/python3.12/asyncio/locks.py", line 212, in wait
    await fut
asyncio.exceptions.CancelledError

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 666, in _execute_local
    resp = await asyncio.wait_for(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/tasks.py", line 519, in wait_for
    async with timeouts.timeout(timeout):
               ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/asyncio/timeouts.py", line 115, in __aexit__
    raise TimeoutError from exc_val
TimeoutError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1206, in _handle_function_tools
    async for resp in self._iter_tool_executor_results(executor):  # type: ignore
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1561, in _iter_tool_executor_results
    yield next_result_task.result()
          ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/agent/runners/tool_loop_agent_runner.py", line 1531, in _next_executor_result
    return await anext(executor)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 185, in execute
    async for r in cls._execute_local(tool, run_context, **tool_args):
  File "/AstrBot/astrbot/core/astr_agent_tool_exec.py", line 699, in _execute_local
    raise Exception(
Exception: tool get-current-date execution timeout after 120 seconds.
[2026-08-28 07:02:05.976] [Core]
[INFO]
[runners.tool_loop_agent_runner:1349]: Tool `get-current-date` Result: error: tool get-current-date execution timeout after 120 seconds.
[2026-08-28 07:02:05.992] [Core]
[INFO]

辅助信息

No response

检查清单

• 我已在 Issue 列表中搜索过相关问题仍无法解决，或该问题从未被报告过。
• 我已尝试过禁用所有插件，排除了可能是因插件导致的问题。
• 我报告的问题与 AstrBot 本体相关，而非在报告某一插件的问题。
• 我已阅读并同意本项目的贡献者行为准则。
• (可选)我愿意提交 PR 以帮助修复该问题。