Python Telegram Bot（v20）突然静默及NetworkError错误处理求助

我用python-telegram-bot v20开发的Telegram机器人偶尔会停止响应，并触发以下错误：

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

Traceback (most recent call last):
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/ext/_application.py", line 1729, in process_error
    await callback(update, context)
  File "/Users/leakyspoon/Documents/TelegramBots/GPT.py", line 301, in error_handler
    await context.bot.send_message(chat_id=492065050, text=message)
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/ext/_extbot.py", line 2644, in send_message
    return await super().send_message(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/_bot.py", line 388, in decorator
    result = await func(self, *args, **kwargs)  # skipcq: PYL-E1102
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/_bot.py", line 814, in send_message
    return await self._send_message(
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/ext/_extbot.py", line 514, in _send_message
    result = await super()._send_message(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/_bot.py", line 566, in _send_message
    return await self._post(
             ^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/_bot.py", line 476, in _post
    return await self._do_post(
           ^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/ext/_extbot.py", line 332, in _do_post
    return await super()._do_post(
           ^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/_bot.py", line 504, in _do_post
    return await request.post(
           ^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/request/_baserequest.py", line 168, in post
    result = await self._request_wrapper(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/request/_baserequest.py", line 288, in _request_wrapper
    raise exc
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/request/_baserequest.py", line 278, in _request_wrapper
    code, payload = await self.do_request(
                    ^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/leakyspoon/Documents/TelegramBots/venv/lib/python3.11/site-packages/telegram/request/_httpxrequest.py", line 240, in do_request
    raise NetworkError(f"httpx.{err.__class__.__name__}: {err}") from err
telegram.error.NetworkError: httpx.ConnectError: [Errno 8] nodename nor servname provided, or not known

请问如何预防该错误，或通过错误处理器进行有效处理？

预防措施 #

• 检查网络稳定性：这个错误本质是DNS解析失败或网络连接中断，先确保机器人运行环境的网络稳定，能正常解析Telegram的API域名（api.telegram.org）。可以定期用ping api.telegram.org或nslookup api.telegram.org验证。
• 配置HTTPX的重试机制：python-telegram-bot v20默认用HTTPX作为请求库，你可以自定义请求对象，添加重试策略来处理临时网络故障：

  from telegram.request import HTTPXRequest
  from httpx import Timeout, AsyncClient
  from httpx_sse import AsyncClient as SSEAsyncClient
  
  # 配置重试策略，针对连接错误、超时等重试3次
  client = AsyncClient(
      timeout=Timeout(10.0, connect=5.0),
      limits=None,
      transport=None,
      retries=3
  )
  sse_client = SSEAsyncClient(
      timeout=Timeout(10.0, connect=5.0),
      limits=None,
      transport=None,
      retries=3
  )
  request = HTTPXRequest(client=client, sse_client=sse_client)
  
  # 在创建Application时传入自定义request
  application = Application.builder().token("YOUR_TOKEN").request(request).build()
  

• 使用代理（如果需要）：如果所在环境无法直接访问Telegram API，配置HTTP代理或SOCKS代理，确保代理服务器能正常解析域名。

错误处理器优化 #

你的现有错误处理器在触发NetworkError时还尝试发送消息，这会导致二次错误（因为此时网络已经有问题），需要修改错误处理器：

• 避免在网络错误时发送消息：先判断错误类型，如果是NetworkError，优先记录日志，不要尝试调用Telegram API：

  from telegram.error import NetworkError
  import logging
  
  logger = logging.getLogger(__name__)
  
  async def error_handler(update, context):
      # 记录错误日志
      logger.error(f"Update {update} caused error {context.error}")
      
      # 只在非网络错误时尝试发送通知
      if not isinstance(context.error, NetworkError):
          try:
              await context.bot.send_message(chat_id=492065050, text=f"错误发生：{str(context.error)}")
          except Exception as e:
              logger.error(f"发送错误通知失败：{e}")
      else:
          logger.warning("网络连接错误，跳过通知发送")
  

• 添加机器人自动恢复逻辑：如果检测到连续多次NetworkError，可以考虑重启机器人的更新轮询（对于Polling模式），或者重新建立Webhook连接（对于Webhook模式）：

  # 示例：Polling模式下的重启逻辑
  async def error_handler(update, context):
      logger.error(f"错误：{context.error}")
      
      if isinstance(context.error, NetworkError):
          logger.warning("网络异常，尝试重启轮询")
          # 停止当前轮询
          await context.application.stop()
          # 重新启动轮询
          await context.application.run_polling()
  

内容的提问来源于stack exchange，提问作者LeakySpoon