fix: issue#4346

[mqk]

What does this PR address?

This PR fixes issue #4346.

Before submitting:

• Does the Pull Request follow Conventional Commits specification naming? Here are GitHub's
  guide on how to create a pull request.

• Does the code follow BentoML's code style, pre-commit run -a script has passed (instructions)?
  The exception is that pdm-lock-check failed, for reason I don't understand. This was a result of using a newer version of pdm (2.11.1 rather than 2.10.4, which your CICD uses). Downgrading pdm solved the issue.

• Did you write tests to cover your changes?

No, but I think it would be a good idea to write a test with a sufficiently large payload that would trigger this error without the fix. But I don't know where or how to do that, so would need some guidance.

[mqk]

Older versions of aiohttp(e.g. 3.8.6) don't expose the ClientSession.max_field_size kwarg.

[mqk]

I've attached an example to reproduce the issue on main. Usage:

> tar xzf picklable_model.tar.gz
> cd picklable_model/
> python ./save_model.py
> bentoml build
> bentoml serve picklable_model_imputation:latest --port 5001

Then navigate to localhost:5001 and click on "Try it out" > "Execute" under /impute. You'll get this error on main:

2023-12-22T11:09:41-0800 [ERROR] [api_server:picklable_model_imputation:9] Exception on /impute [POST] (trace=238a2c870205053cf22121983fc3cb55,span=a2da156f8a15101d,sampled=0,service.name=picklable_model_imputation)
Traceback (most recent call last):
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/client_reqrep.py", line 905, in start
    message, payload = await protocol.read()  # type: ignore[union-attr]
                       ^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/streams.py", line 616, in read
    await self._waiter
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/client_proto.py", line 213, in data_received
    messages, upgraded, tail = self._parser.feed_data(data)
                               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "aiohttp/_http_parser.pyx", line 557, in aiohttp._http_parser.HttpParser.feed_data
  File "aiohttp/_http_parser.pyx", line 732, in aiohttp._http_parser.cb_on_header_value
aiohttp.http_exceptions.LineTooLong: 400, message:
  Got more than 8190 bytes (16293) when reading Header value is too long.

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

Traceback (most recent call last):
  File "/Users/mikekuhlen/Code/BentoML/src/bentoml/_internal/server/http_app.py", line 343, in api_func
    output = await run_in_threadpool(api.func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/starlette/concurrency.py", line 41, in run_in_threadpool
    return await anyio.to_thread.run_sync(func, *args)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/anyio/to_thread.py", line 33, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/anyio/_backends/_asyncio.py", line 2106, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/anyio/_backends/_asyncio.py", line 833, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/bentoml/bentos/picklable_model_imputation/uasnkvfa7w5ctaty/src/service.py", line 16, in impute
    return imputation_runner.run(df)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/src/bentoml/_internal/runner/runner.py", line 52, in run
    return self.runner._runner_handle.run_method(self, *args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/src/bentoml/_internal/runner/runner_handle/remote.py", line 346, in run_method
    anyio.from_thread.run(
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/anyio/from_thread.py", line 45, in run
    return async_backend.run_async_from_thread(func, args, token=token)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/anyio/_backends/_asyncio.py", line 2121, in run_async_from_thread
    return f.result()
           ^^^^^^^^^^
  File "/opt/homebrew/Cellar/[email protected]/3.11.5/Frameworks/Python.framework/Versions/3.11/lib/python3.11/concurrent/futures/_base.py", line 456, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/opt/homebrew/Cellar/[email protected]/3.11.5/Frameworks/Python.framework/Versions/3.11/lib/python3.11/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/Users/mikekuhlen/Code/BentoML/src/bentoml/_internal/runner/runner_handle/remote.py", line 216, in async_run_method
    async with self._client.post(
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/client.py", line 1167, in __aenter__
    self._resp = await self._coro
                 ^^^^^^^^^^^^^^^^
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/client.py", line 586, in _request
    await resp.start(conn)
  File "/Users/mikekuhlen/Code/BentoML/.venv/lib/python3.11/site-packages/aiohttp/client_reqrep.py", line 907, in start
    raise ClientResponseError(
aiohttp.client_exceptions.ClientResponseError: 400, message='Got more than 8190 bytes (16293) when reading Header value is too long.', url=URL('http://127.0.0.1:8000/')
2023-12-22T11:09:41-0800 [INFO] [api_server:picklable_model_imputation:9] 127.0.0.1:49864 (scheme=http,method=POST,path=/impute,type=application/json,length=83358) (status=500,type=application/json,length=110) 860.815ms (trace=238a2c870205053cf22121983fc3cb55,span=a2da156f8a15101d,sampled=0,service.name=picklable_model_imputation)

picklable_model.tar.gz

[jianshen92]

Thanks for opening this PR. I believe this is resolved on the data container level in this PR . I will close this one