Re:A Mechanised Cryptographic Proof of the WireGuard VPN Pro...
by cyphar on Monday May 13, 2019 @07:32AM
(#19898168)
It is possibly the most usable VPN I've used (in terms of resilience -- configuration is pretty easy but could be made more out-of-the-box ready).
Almost every distribution has packages for it[1] and it's just a matter of copying the example wg-quick configurations from the internet. Personally I've written a more complete script[2] that allows you to dynamically add new devices to a server (and generate a QR code for the Android client).
Yes, it's not mainline yet but it appears to be getting quite close and people still use VirtualBox even though its drivers are out-of-tree. Even though Donenfeld hasn't started tagging releases that are CVE-eligible I'm willing to bet that WireGuard is more secure than its alternatives (has anyone even attempted a formal proof of OpenVPN or IPSec?).
Almost every distribution has packages for it[1] and it's just a matter of copying the example wg-quick configurations from the internet. Personally I've written a more complete script[2] that allows you to dynamically add new devices to a server (and generate a QR code for the Android client).
Yes, it's not mainline yet but it appears to be getting quite close and people still use VirtualBox even though its drivers are out-of-tree. Even though Donenfeld hasn't started tagging releases that are CVE-eligible I'm willing to bet that WireGuard is more secure than its alternatives (has anyone even attempted a formal proof of OpenVPN or IPSec?).
There are a few userspace implementations too.
[1]: https://www.wireguard.com/install/ [2]: https://github.com/cyphar/dotfiles/blob/master/.local/bin/wg...