Hm yeah, the problem is that we've locked down the server the relay ran on. Port 47474 is open for the outside to connect to, but the relay is not allowed to make outgoing connections to random ports (the hosts). This is due to us locking traffic down upstream, on a switch and not a firewall. And the switch isn't stateful, so it's not capable of discerning between incoming and outgoing connections / traffic.
Because of that, I've moved the relay to another server with some more networking freedom. The relay is now connecting to all hosts again and you can use it if you connect [your app] to isrelay.lfs.net .
BUT this brings another problem ... the LFS Remote apparently was coded to connect to "lfs.net" and not "isrelay.lfs.net". So now it's trying to connect to the wrong thing. On top of that I seem to have displaced the source code for the LFS Remote ... so now there is no way of fixing it.
Sooooo ... I guess the need for a HTML5 remote is really there now.
It's slightly related to the attacks, but not directly. I know what the problem is, but I haven't had the time to fix it (the post indexer isn't currently running, so new posts are not indexed).
Will find some time when the current issues have gone.
Because you did not use the uninstaller, some registry entries have remained. That makes the installer think you still have installs there.
I think you can find the registry entries in HKEY_CURRENT_USER\Software\Live for Speed
You'll probably have multiple of those. Delete any you don't have anymore. You can see to which LFS folder they belonged by viewing all their details.
oh, yeah there may also still be one or more Windows Start Menu items that you may have to manually remove.
ah, lfsforum.net hasn't been moved to CloudFlare and so all requests go directly to our web server, but you'll be blocked there (no direct requests allowed anymore)
can move that domain as well.
Do you still use it then?
We try to null-route (plainly block) attacks, preventing servers from being drowned in traffic so that other LFS services remain available, which is why you could still ping it. But by blocking the attacks (closing ports), we do block the service behind it as well, which is why the master is / was down from time to time.
The websites now sit behind CloudFlare and the web server has been locked down with switch ACL's. This already has protected the website from attacks late last night.
The website indeed has always been available through IPv6 because the attack only occurred on IPv4.
Finally, we have a plan to better protect the master, however it will take a bit of work and a patch for LFS. Hopefully that will come later this week.