I'm trying to make wireguard work with a special config, to test Browsers for WebRTC Leaks. With this said I need a config, that allows allows traffic outside of the tunnel. In Windows I could accomplish this by turning of the "Block untunneled traffic (kill-switch) checkbox" on the client, but the config file I get from this (see below) isn't working on Linux (ubuntu) with wg-quick, because no working tunnel is created.
I saw suggestions for splitting tunnels but I'm not shure how to adress this here since I don't want so spilt the tunnel for a specific Server or IP-Range, but allow WebRTC to also route over the Default Gateway (which would make the Leak visible).
[Interface]
PrivateKey = key
Address = 10.14.0.2/16
DNS = 162.252.x.x, 149.154.x.x
[Peer]
PublicKey = fJDA+OA6jzQxfRcoHfC27xz7m3C8/590fRjpntzSpGo=
AllowedIPs = 0.0.0.0/1, 128.0.0.0/1
Endpoint = example.com:51820
Any ideas, why this isn't working on Linux? Or maybe ideas how I can set up this usecase in an Linux-Environment? When I set AllowedIPs = 0.0.0.0/0 in the Linux-Config, then the tunnel works, but now all the traffic is routed through the tunnel (WebRTC Leak not visible).
Maybe the setup is a bit special but I would really like to understand why this isn't behaving in the same way. I would really appreciate your answer.