Thanks for that! I hadn't heard, before, that anyone had successfully used the "Birthday Paradox" on port numbering to successfully defeat NATs. Glad to see t

Thanks for that! I hadn't heard, before, that anyone had successfully used the "Birthday Paradox" on port numbering to successfully defeat NATs. Glad to see that Tailscale has successfully implemented and tested this -- not just at one layer, but at multiple layers of NATs!