[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <440F267B.80703@uniontown.com>
Date: Wed Mar 8 18:50:14 2006
From: securitylists at uniontown.com (Security Lists)
Subject: Re: recursive DNS servers DDoS as a growing
DDoSproblem
Sorry, I don't see this as amplification in your example, because YOUR
dns servers are 100% of the traffic. 1:1 ratio.
Now, if you get the world to cache your text records, and have THEM
flood with source-spoofed UDP (unrelated to the victim's DNS servers),
that'd work, and is actually a good example of amplification. There's
plenty of open DNS servers out there that'll do this, but each one of
them is going to hit your local dns server initially.
-Mark Coleman
Geo. wrote:
>>> In the scenario you describe, I cannot see any actual amplification...
>>>
>
> I'll give you a senario where you can see.
>
> lets say you have 2 name servers that are local to you.
>
> I setup a domain, example.com. In this domain I create a text record which is 100K in length, I don't know, perhaps I paste the source code to decss in it, whatever it's a big text record.
>
> Now I simply spoof a UDP packet using your IP address as the source address and send it to both of your dns servers. This packet is a query for the example.com text record. I have now sent two very small packets and you have received 200K of traffic. That's the amplification, one small udp packet, one large text record in return.
>
> Note, I don't have to use your local servers, but this way it makes it more fun to troubleshoot because it looks like you are the cause of your own flooding..
>
> Geo.
>
> _______________________________________________
> Full-Disclosure - We believe in it.
> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
> Hosted and sponsored by Secunia - http://secunia.com/
>
>
Powered by blists - more mailing lists