Unhandled exception ConnectionResetError: Cannot write to closing transport

[appliedprivacy]

doh-httpproxy running on 9f943a4
we got this in the logs:

 ERROR: Unhandled exception
 Traceback (most recent call last):
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/web_protocol.py", line 447, in start
     await resp.prepare(request)
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/web_response.py", line 353, in prepare
     return await self._start(request)
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/web_response.py", line 667, in _start
     return await super()._start(request)
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/web_response.py", line 410, in _start
     await writer.write_headers(status_line, headers)
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/http_writer.py", line 112, in write_headers
     self._write(buf)
   File "/usr/local/lib/python3.7/dist-packages/aiohttp/http_writer.py", line 67, in _write
     raise ConnectionResetError('Cannot write to closing transport')
 ConnectionResetError: Cannot write to closing transport

[chantra]

Thanks @appliedprivacy Can you confirm that this is just a logging cosmetic issue and does not prevent the proxy from working correctly? What http proxy is running in front of DoH-proxy (e.g nginx for instance)? What is the relevant config of the http proxy? Anything specific that triggers this error? Like slow recursive query, or a restart of the http proxy for instance? From looking at the logs, it seems to indicate that DoH-proxy is writing a response back to the http proxy, but the proxy has closed the connection in between sending a request and getting the response. This could be explained by either the http proxy deciding that the request took too long to get a reply (some sort of timeout ) or that the http proxy was killed/restarted between the time DoH-proxy receive the request and the reply is sent back. In any cases this is something that would need to be handled in http_proxy.py most likely.

[appliedprivacy]

Can you confirm that this is just a logging cosmetic issue and does not prevent the proxy from working correctly?

yes, I think so

What http proxy is running in front of DoH-proxy (e.g nginx for instance)?

nginx

What is the relevant config of the http proxy?

it matches more or less your guide here:
https://facebookexperimental.github.io/doh-proxy/tutorials/nginx-dohhttpproxy-unbound-centos7.html

Anything specific that triggers this error? Like slow recursive query, or a restart of the http proxy for instance?

I don't know, but nginx did not restart at the time of the error log.
This error did only show up once in the logs.

I'll keep an eye on it and let you know should it happen again.

[chantra]

the error is raised here:
https://github.com/aio-libs/aiohttp/blob/master/aiohttp/http_writer.py#L67

Basically when we try to write data back to the HTTP client, the transport had already been closed.

At this stage there is nothing really actionable, and I don't think it is worth hiding the exception. We could spray the code with checks whether the transport is not in closing state but I dont think it is worth it.

I will close this issue for now, if this was becoming a real issue, please open a new issue.