Re: [NSE] http-slowloris, check if a webserver is prone to the Slowloris DoS attack

[Henri Doreau <henri.doreau () greenbone net>]

2011/6/12 Gutek <ange.gutek () gmail com>:
> So where do I put the global cursor ?
> This would require asking the user about the presumed weakness of his
> server. For example, if he considers it "weak", then a 10 minutes max
> attack would be sufficient to state about this vulnerability. But if he
> considers it "strong", the script would have to run maybe a day long to
> be sure. But this means defining "weak" and "strong" in terms of
> numbers. Not speaking about "blind" conditions when testing an unkown
> target.
> On the other hand I agree that the attack can not last for ever. I just
> can't say "how" (in fact, "when") stop it.
Ok, as you understood I meant "give up if the server is still alive".

The attack will reach a stable state (max number of alive connections)
after a while. Wouldn't that make sense to give up and consider that
the server is not vulnerable if the target is still alive at this
point? It wouldn't mean that the target is not vulnerable to slowloris
attacks, but it not with the selected
--max-parallelism/MAX_ATTACK_THREADS combo.

> don't know why, but this famous quote comes to my mind "This is not
> mission difficult, Mr. Hunt, it's mission impossible" :)
>
> A.G.
Actually I think that our Ethan Hunt (Patrik Karlsson) got it right,
if you wait for all the worker threads to be done before you exit the
main thread you get rid of that error. Can you fix this? (there is an
example in stdnse.new_thread documentation).

Regards.


-- 
Henri Doreau |  Greenbone Networks GmbH  |  http://www.greenbone.net
Neuer Graben 17, 49074 Osnabrueck, Germany | AG Osnabrueck, HR B 202460
Executive Directors: Lukas Grunwald, Dr. Jan-Oliver Wagner
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/