I am experimenting with using forgejo instead of GitHub for my personal projects. So far I like it, however I would like to make it available to the outside world at some point.
I was wondering what kind of traps I should avoid. The following things come to mind so far:
- Forgejo Actions seem like a massive potential security risk, however I do not intend to enable sign up for other
- OpenID appears to be a thing for forgejo, I do not know how it works and it seems like it would allow access to my instance even with registering disabled
- I would put the instance behind a nginx as reverse proxy, but how do you keep bot traffic to a minimum? Anubis?
I feel like there are a ton of things I have not thought of, which is why I am holding off on making anything available without a VPN so far.
I run a forgejo server on the internet for myself so here’s my two cents.
OpenID or any other log in method is just a way to log in to an account on the forgejo server. With registrations disabled it shouldn’t be able to create an account, so there should be no issue. All the extra auth methods also need to be configured and are disabled by default.
Bot traffic can be pretty horrendous once they find your server, I’d say Anubis is basically a requirement at this point. I would also strongly recommend setting up fail2ban for http and ssh, and disabling ssh password auth.
Ah I see, I kinda assumed that OpenID would allow anyone with an OpenID server to just log in. That seemed like it would cause immediate spam issues. It would definitely be nice if I could allow users to create issues or stuff like that at some point. I will definitely have to take a look at fail2ban before I make it available. Thank you.