Alexander Mohr

@zferentz that's totally normal, you're not going to have all your code async, only that which makes blocking calls, or eventually makes blocking calls.

@zferentz I don't see any blocking io calls, can you point that out?

@zferentz ya but there are no python built-in async functions for file io, we'd have to swap to using aiofiles or manually via threading which is non ideal. There really...

@zferentz it is cached, at the session/client level :), and people really shouldn't be re-creating sessions/clients often as there's no need. In terms of async code calling blocking io yes...

FYI to track slow sync calls you can use something like this: https://gist.github.com/thehesiod/0d2beed27c800e4f7286be6c3cd17a09 That's what we have in production to ensure sync calls aren't blocking the main thread for too...

@zferentz I think it would be great to speed up botocore, actually I proposed some changes to how they parse timestamp years ago that sped up our scenario 20% and...

and in either case can use your monkeypatch to speed up your use case :). Actually if this really does take a significant amount of time I'd like to use...

validated that the body is getting closed in aiohttp 2.x

due to aiohttp 2.x change

have submitted a provisional fix for this until we have a final resolution on the aiohttp issue I logged