Maybe still issues with test_large_io_doesnt_hang on windows?

[bollwyvl]

trafficstars

Seeing some test fails on https://github.com/conda-forge/terminado-feedstock/pull/38, looking into it:

        if future_cell[0].cancelled() or not future_cell[0].done():
>           raise TimeoutError("Operation timed out after %s seconds" % timeout)
E           asyncio.exceptions.TimeoutError: Operation timed out after 30.0 seconds

..\_test_env\lib\site-packages\tornado\ioloop.py:528: TimeoutError
------------------------------ Captured log call ------------------------------
WARNING  tornado.general:autoreload.py:128 tornado.autoreload started more than once in the same process

https://dev.azure.com/conda-forge/feedstock-builds/_build/results?buildId=575370&view=logs&j=7df5a329-fcdb-5857-5720-a5dae94c0641&t=428acfd7-2921-5219-037f-9169d136e406&l=494

Maybe i missed some test dep changes? Might be the weekend before i can investigate further...

[bollwyvl]

Yep, was missing pytest-timeout: now it's just win/pypy38 and win/cpython39, so might be flake:

https://dev.azure.com/conda-forge/feedstock-builds/_build/results?buildId=575374&view=logs&j=e5cdccbf-4751-5a24-7406-185c9d30d021&t=66e4a587-a723-53e9-873c-6c9a2c5971ea&l=499

[KoopaKing]

Ah, yeah, I may have tuned that timeout too tightly... on my system that test was taking around 20 seconds, but I didn't want it to take forever on failures.

[blink1073]

Builds are passing on conda-forge now, I think we're good!