McMonster

joined 2 years ago
[โ€“] McMonster@programming.dev 3 points 1 day ago (1 children)

Thanks. Plain Wireguard is an option I'm considering, but it's also considerably more hassle to configure and maintain, especially as I connect more family members to my network. Headscale also has an extra layer of security in the form of ACLs, which I plan to use on top of basic firewall configuration. I do connect my personal machines with Wireguard, but I use one family member as a Tailscale/Headscale test subject.

As for SELinux, I've gave up on it already. It caused me so much headache over the years I disable it with a kernel parameter by default on all machines.

 

Did anyone paranoid like me research security implications of running Tailscale/Headscale or similiar?

Right now I'm self-hosting headscale controller in my LAN and expose it to public Internet. I'm thinking about moving it to a VPS, but I'm a little paranoid about exposing the software that controls connectivity between my and family machines to a third party, be it official Tailscale controller or VPS provider where I run Headscale.

Currently I think that even in the worst case of someone compromising my Headscale instance it should still be fine as long as all of the machines are properly firewalled and all of the exposed apps and services are behind authentication. I run everything behind Authentik and only keys for SSH access. I will certainly add some network monitoring to all of that.

Any opions and suggestions on this matter are welcome.

The same also applies to other things, like tools.

I'm working in Java ecosystem and there's a noticeable trend to look down on people who don't use IntelliJ Idea. I've recently joined a new project and I was strongly encouraged to use it. Therefore I'm currently 3 weeks into my 4th attempt over past 10 years to switch to this tool and it simply doesn't work for me. I've been using Eclipse since around 2007, know it very well and it gets the job done. I will not claim that it is better than Idea. I just don't think switching would give me enough return on investment. Especially now, as I'm still learning the new project.

Another reason not to switch is to avoid becoming dependant on an expensive tool. My current team is using Ultimate edition and I've noticed that they are really depending on the extra features.