The bad news is that denial-of-service (DoS) attacks are becoming
more numerous on the Internet. Not only are DoS attacks more frequent, they are
more potent with the potential to do much greater harm than they've done to date.
The good news? Right now, according to experts, there isn't any.
DoS attacks overwhelm computers, Web sites and servers with floods
of bogus data , and hackers are increasingly aiming them at routers,
according to a recent report by the federally funded
Computer Emergency
Response Team (CERT). Routers are the vital Internet components,
either special-purpose computers or software packages, that connect two
or more networks or parts of networks.
"Essentially routers have trust relationships with each other, and
are the means by which networks interconnect with each other," Kevin
Houle, one of the authors of a CERT white paper on the subject, told
NewsFactor Network.
"If I can take advantage of that trust relationship
to inject bogus routes in the routing tables, there's a potential for
denial-of-service between two or more networks. They can be separated
from each other."
Massive Traffic Jam
Routers do not have monitoring technology -- they spend their time
looking at the destination addresses of the data packets passing through
them and determining which route to send them on. Routers are the keys to
larger networks, and if they are isolated, considerable disruption could
occur on the Internet.
"Traditionally, you think of DoS as 'packet flooding,' sending
enough traffic down a pipe to fill up that pipe," Houle said. "In the
case of a router-based DoS attack, what we're talking about is the route
tables for a router being altered."
A targeted attack that shut down a network router would not bring
the entire Internet to a halt -- it would be more like a massive
rush-hour traffic jam on an interstate highway that once flowed
smoothly.
'Autonomous Network Worms'
The CERT research also found that multiple-source attacks are
occurring more often and are increasingly aimed at multiple targets.
"Autonomous network worms" are becoming more popular among the more
sophisticated, malicious users, whereas once they simply inserted code
manually via a Trojan Horse into the targeted computer.
"In the case of the automatic model, the attack code is
self-contained," Houle said. "In previous worms like
ramen, the attack
code was in an external site. The compromised computer had to go back to
the attacking host to retrieve a copy of the attack code, install it and
then execute it. The autonomous model is much more efficient. It doesn't
have to take as many steps to initiate another attack." (continued...)
|