The next generation of DoS attacks

[edesignuk]

/. Article:
An anonymous reader writes "A paper from Rice University appearing at the 2003 ACM Sigcomm Conference presents a new denial of service attack where the attacker only needs to send at a low rate to shutdown TCP flows. The trick exploits the retransmission timeout mechanism in TCP. By sending small bursts of packets at just the right frequency, the attacker can cause all TCP flows sharing a bottleneck link to simultaneously stop indefinitely. And because the attacker only needs to burst periodically, the attacker will not be distinguishable from normal hosts. The presentation, and other presentations from the conference, are available online (live streaming)."

Link.

 

[Mr. Anderson]

Ah, now that's going to cause real problem. Papers like this are just fantastic - why don't you go and tell the hackers what to attack next or how...

D

 

[edesignuk]

Originally posted by Mr. Anderson
Papers like this are just fantastic - why don't you go and tell the hackers what to attack next or how...

My thought exactly. "Here, we have found a great new way to take sites down without the need for masses of bandwidth, wanna know how? Here let me explain..."

 

[MrMacMan]

Yes...

Lets Destroy those computers...

Slowly...



What the hell... this takes all the hard work out of DoSing a person!


No more mass gathering to take the RIAA down?

 

[Mr. Anderson]

Originally posted by MrMacman
Lets Destroy those computers...

Slowly...



What the hell... this takes all the hard work out of DoSing a person!

I don't think you get it - its not about doing it slowly, its more like a stealthy approach that makes it very hard to find and counter. This is really serious and the effect would be the same as if it was a regular DoS attack.

D

 

[patrick0brien]

-edesignuk

I have a general problem with this releasing of information. The biggest problem is that I'm torn that it should or not. But in this particular case, it causes me to think it's similar to publishing information on a easy-to-build nuke out of matchsticks, a capacitor, firecrackers and that sphere of beryllium you just happen to have lying around.

This should be distributed to those who can solve the issue long before the public can get their hands on it.

It also reminds me of "Boy this stinks, here, take a whiff..."

 

[Ambrose Chapel]

i guess it's a race...now that the info is out there for everyone, who will use it first - those who will use it to hack and those who will use it to patch?

 

[CMillerERAU]

I suppose the good side of letting all this out is hopefully it will drive IT folks to develop solutions to this problem.

 

[MrMacMan]

Well they need to find out *how* to do this...

And see if it works on every OS and stuff.


I really don't understand how a lower powered version works...



With normal ones you overwhelm the computer... with this... it seems different...

 

[themadchemist]

Originally posted by Mr. Anderson
Ah, now that's going to cause real problem. Papers like this are just fantastic - why don't you go and tell the hackers what to attack next or how...

D

the grad students who write these papers probably don't mind that all too much...

They could also be hackers, living a double life, one as an academic hack and the other a real-world hacker...

That, or they'd like to have the satisfaction of saying, "Yep. My paper led to this massive attack! w00t!" (An impending PhD does not prevent the excessive and inappropriate use of the word w00t)

 

[patrick0brien]

Originally posted by MrMacman
I really don't understand how a lower powered version works...

-MrMacman

Think of it analogous to a slow and steady pressure with a tip of a pen, rather than a deep and fast stab of a knife.

 

[iDONTsteal]

Should?

As soon as we yield the power of what should be published to <insert power mongloid here>, we have begun our way down an even worse path.

Plus... copyrights allow the spread of information and ideas while protecting their <perhaps too long> profit potential...


There are certainly other examples... but.. information spread is good... in almost every case