As a reminder from our discussion yesterday, we want to move this into sshd/enter.py with a .wait() to detect when the container stops, and sending a SIGSTOP/SIGKILL or whatever to the docker exec command. There's a few edge cases around that.

I switched the implementation, but it's giving some odd behavior.
Killing the docker exec causes ssh tests to fail, currently they all pass but that's only because I only kill if it's a kata container and we don't test with any of those. It seems to work fine locally I'm not sure why killing the exec causes the tests to fail.
Another thing is that whenever the docker exec is killed there is be extra white space before the prompt shows up.

@ConnorNelson I believe this is good, it works locally

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!