Forum
Any easy ways to avoid the ddos?
Created 5th August 2009 @ 08:44
Add A Reply Pages: « Previous 1 2 3 Next »
zBlock has nothing to do with a Distributed Denial of Service attack.
lol what? how can you say that.. do you know what you’re talking about?! Zblock stops alot of common/easy to find out ways of DDOS’ing a source server…
zBlock has nothing to do with a Distributed Denial of Service attack.
lol what? how can you say that.. do you know what you’re talking about?! Zblock stops alot of common/easy to find out ways of DDOS’ing a source server…
Yes sir, I do know what I’m talking about and no, Zblock can in _NO POSSIBLE WAY_ block _ANYTHING_ that has REMOTELY to do with DDOS :)
Here is an simple white paper from its conception (to keep it simple and basic), if you wish to learn more on the subject: http://www.linuxsecurity.com/resource_files/intrusion_detection/ddos-whitepaper.html
That said, the only way to keep matchservers some-what “safe” is to:
a) Make them not register themselves
b) Make sure it’s on a relatively unknown host and use a random port together with something like portsentry on the hosts that runs on the backend, so they can’t find out where the game is played by portscanning (and getting blocked if they try).
c) Every time games of high attention are played, all participants set their steam-friends to show offline before connecting to the server.
d) Mask the STV from showing the originating IP via “status”.
It’s truly sad it must be done this way, but it’s the only way.
You’re wrong, coming from css I know you’re wrong because the game was plagued with the attacks untill a few releases ago.
YOU ARE AWARE that when people go round saying omgofmgfogmf they ddos’d the server its very rarely an actual DDOS yeah? It’s just some shitty commands, script or whatever that fucks with the source engine.
You’re suggestions are WHEY WHEY WHEY OTT. Just get in touch with J3di and get a TF2 port of Zblock on the go. Seriously.
You’re wrong, coming from css I know you’re wrong because the game was plagued with the attacks untill a few releases ago.
YOU ARE AWARE that when people go round saying omgofmgfogmf they ddos’d the server its very rarely an actual DDOS yeah? It’s just some shitty commands, script or whatever that fucks with the source engine.
You’re suggestions are WHEY WHEY WHEY OTT. Just get in touch with J3di and get a TF2 port of Zblock on the go. Seriously.
Pretty much, most rage kiddies wouldn’t know how to proplerly fuck up a server which is why zBlock works.
Why do I even bother… Fine, just because I’m such a nice and FRIENDLY guy! :D
The ACTUAL concept of DDOS is and has been a real “threat” to games, by packeting the gameservers themselves.
If you’d like to call crashing a server by executing a bunch of… say… common Q-Engine flaw, a bunch of (thousands) ‘ý’s on its port via some UDP packet(s) a DDOS-attack, that’s just plain wrong.
If you’d say it was a Denial Of Service attack, sure — but both issues are a huge problem for any high-profile service and are dealt with differently.
Nobody likes you
coming from nvc, I kinda lold :[
also just figure out from the server logs if it actually was some source engine abusing or a real ddos. Abusing can be patched and whatever but DDOS can only be “stopped” with completely hiding the server where the game is played.
you might as well whitelist the IPs of the players actually connecting to the server, including 1 stv relay.
Wouldnt that work?
the packets coming to the “firewall” block the connection anyway. whitelisting doesn’t help
you guys need to understand the mechanics of ddos :p
you might as well whitelist the IPs of the players actually connecting to the server, including 1 stv relay.
Wouldnt that work?
It would work just as well as zBlock, but in a broader perspective, yes. This requires blocking done via IP-filtering though, not any ordinary “built-in” lists (I do not even know of any such things present in Source).
What I _do_ know by personal experience, is that adding offenders IP to the “banned IP-list” is not enough for them to “pre-execute” commands, much less crash the server itself by sending bogus data to the port. It has to be done as described above to be effective.
… And after all said and done, the suggestion with “hidden” server(s) would solve all these issues in one go, so it’d be the “best” way of doing things.
it’s not too hard to gather a botnet in a couple of hours, so everyone with minimal knowledge of these technologies can ddos any serv. it’s way more easier than some of you think
Add A Reply Pages: « Previous 1 2 3 Next »