Ok that's good to know, thanks for the heads up. How far are you through that testing refactor? I'm assuming it's not something we could likely ship for the cut, given you said the word "refactor" 🙂 I'm wondering then... If there's not some other good, relatively reliable mechanism to verify our exist are working correctly, to the extent that we could be comfortable shipping the new shutdown logic and then have a known issue or something that we need to land your re-arch? Do you have any thoughts on how we could get to a state where you're comfortable with that landing?