Maximilian Hils

For the record, the problem here is that we currently keep flows in memory until the associated connection is closed. My workaround in https://github.com/mitmproxy/mitmproxy/issues/4786#issuecomment-928002790 makes sure that this happens after...

Quick Update: We've just shipped 8.1.0, which greatly alleviates memory issues and renders the workaround posted above superfluous. There are some final parts that depend on a new cryptography release...

That's a fair complaint. One reason why implementing this isn't too easy is that custom add-ons may add additional options, so we need multiple passes here. One easier step would...

> I don't even know were to begin writing tests. What you're saying is I shouldn't use `from multiprocessing import Process` to spawn a process and test the exit code,...

I don't have the time do work on this at the moment, but if someone is interested, wrapping `proxy.ServerConnection` should do the job.

Not yet, sorry.

@bratao: A good first step would be to simply rewrite the existing HTTP Upstream Proxy Implementation with a SOCKSv5 one. Our proxy core is sans-io, which means that python-socks may...

Closing this for a lack of information. :)

There's no functionality for that at the moment, but it's definitely something we could add. :)

I have no plans to work on this myself anytime soon. If someone wants to give it a stab, I'm happy to review a PR.