Image
 
Latest Additions
 
EH-Net Login
Welcome Guest.






Lost Password?
No account yet? Register
Who's Online
We have 17 guests and 4 members online
EH-Net Donations

Enter Amount:
$

Google Ads
ChicagoCon 2008f
chicagocon2008f_125x200banner.jpg
ChicagoCon 2008f
EH-Net News Feeds
Latest Additions
Book Recommendations





 
Advertisement

You are here: Home arrow Forum arrow Ethical Hacking Discussions and Related Certificationsarrow Malwarearrow Hacking: 802.11 Protocol Attacks, Deauthentication
Ethical Hacker Community Forums
September 05, 2008, 04:35:04 PM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: Registration Now Open for ChicagoCon 2008f Oct 27 - Nov 2! Visit www.chicagocon.com.
 
   Home   Help Calendar Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Hacking: 802.11 Protocol Attacks, Deauthentication  (Read 2772 times)
0 Members and 1 Guest are viewing this topic.
bojan
Newbie
*
Offline Offline

Posts: 19


View Profile
« on: March 26, 2008, 02:12:19 AM »

Written by Devin Akin
Tuesday, 19 February 2008

This article is presented as part of hacking + solution track for Wireless Security Expo 2008. The hacking video is available here hxxp://www.cwnp.com/learning_center/video_deauth.html

Deauthentication is the most common form of 802.11 protocol denial-of-service (DoS) attack. After watching the Deauthentication video, you can see that performing this type of attack takes seconds using common and user-friendly software and hardware, can wreak havoc on a network, and can be used as part of other types of wireless network attacks. Deauthentication frames are considered notifications, not requests, which means any associated station or AP that receives a deauthentication frame must comply.
802.11 stations must authenticate themselves through "Open System Authentication" prior to requesting a connection. Following successful authentication (consisting of two acknowledged authentication frames), the client station will then request association (connectivity). The association request frame is followed by an association response frame. Each of these frames are also acknowledged.

The next steps depend on the type of security in use on the WLAN and determine just how intrusive a deauthentication attack will be. If the WLAN is using only Open System authentication, then a deauthentication attack will yield a very minor interruption for client stations. The reason for this is that the authentication and association process is extremely fast. When deauthenticated, a client station must reauthenticate and reassociate, but this entire process takes only a few milliseconds to complete. If the WLAN is using WEP with Open System authentication, the same process would apply.

If the WLAN is using WPA/WPA2-PSK, then a 4-way handshake (plus 4 ACK frames) will follow the acknowledged association response frame. This process is fairly fast (roughly an additional 20-30 ms), but added to the Open System authentication and association, it can easily add up to 50 ms (total) when adding in contention time. If a single AP is used, this won't be a big problem, but a deauthentication like this may also cause a client station to roam. Roaming requires passive and active scanning, which could add 1-3 seconds to the process. This additional time can easily disrupt many applications.

If the WLAN is using 802.1X/EAP and not using Opportunistic PMK Caching (not widely supported in client utilities), deauthentication can cause a disruption of 0.5 - 5 seconds depending on the specific EAP type in use, scanning processes, and the 4-way handshake. 802.1X/EAP authentication mechanisms are almost always deployed in enterprise WLANs. Any application that is latency sensitive will suffer dramatic problems when the client station is deauthenticated. File transfers, voice/video streams, thin-client sessions, and other real-time applications will often break when disrupted for more than 0.5 seconds.

The 802.11w amendment to the 802.11-2007 standard offers three new security pieces: Data Origin Authenticity, Replay Detection, and Management Frame Protection. The data origin authenticity mechanism defines a means by which a station that receives a management frame (such as a deauthentication frame) can determine which station transmitted the data or management frame. This feature is required to prevent an intruder from masquerading as an authorized station. The replay detection mechanism defines a means by which a station that receives a management frame from another station can detect whether the received frame is an unauthorized retransmission. Management frame protection is required to protect against forgery and eavesdropping on management frames such as Action, Disassociate, and Deauthenticate frames through the use of security keys.

Most of today's WLAN infrastructure systems do not support management frame protection, and until they do, deauthentication attacks will remain a significant security problem.
Logged
bojan
Newbie
*
Offline Offline

Posts: 19


View Profile
« Reply #1 on: March 26, 2008, 02:13:30 AM »

I frgot to say this is my first post in EH-NET!!!

Logged
bojan
Newbie
*
Offline Offline

Posts: 19


View Profile
« Reply #2 on: March 27, 2008, 03:59:38 AM »

but i want 2 ask which is more powerfull in this ddos area??

protocol attacks or a stronge botnet?Huh??
Logged
dean
Full Member
***
Offline Offline

Posts: 130


View Profile
« Reply #3 on: March 27, 2008, 09:35:30 AM »

I will assume that when you say 'protocol attacks' that you mean attacks against the management frames in the 802.11 protocol.

The excerpt you posted talks about the effect of deauth or auth attacks. It's trivial to run a sustained deauth flood on an AP or client effectively keeping them off the network. Note that this is an attack against the local (W)LAN.

A botnet is generally, although somewhat seldom now, going to be used to attack the internet facing presence of an organization. Depending on the size of the botnet and the type of DDoS employed this can be very effective in generating Gbits of traffic to the targets. Consider having a botnet of 100 machines each with only 56k of upstream bandwidth. If these are all directed to DoS a single target a simple SYN flood can cause a significant disruption in traffic.

So to answer your question they are both attacks against services using or abusing the protocols used for communications. They are different in that one can be done remotely and the other requires you to be within range of the AP or client. (and yes, this includes using an amplifier, antenna, etc...)

Is one more severe than the other? Well that depends on which service is more important to the organization.
dean
Logged

<script>alert('%52%54%46%4D')</script>
bojan
Newbie
*
Offline Offline

Posts: 19


View Profile
« Reply #4 on: March 29, 2008, 01:09:40 AM »

Like Devinator says "Most of today's WLAN infrastructure systems do not support management frame protection" .

And that is because it is fairly new (about 4-5 months old ?).

So, maybe if you look at the Management Frame Protection project on ieee, you'll know what all management frame protection is targeting. Don't know if they have an RFC for Management Frame Protection. (Lazy to search .. just woke up  )

Deauthentication is just one type of frame that is shown in the video. But other frame types are certainly more destructive. A couple of the other frame types are mentioned in his paper. But not elaborated.

The best place to try all this would be your own SOHO WLAN or at home on that irate house mate that knows about ACLs and has equal access to the router config on a shared DSL connection
Logged
bojan
Newbie
*
Offline Offline

Posts: 19


View Profile
« Reply #5 on: March 29, 2008, 01:55:14 AM »

A DDoS is possible, but with machines within the vicinity which is possible (if it is within the vicinity, there is a good chance that you actually own these machines), I mean .. what are the odds of finding loads of hosts and that too ones using 802.11 ..

What further reduces the chances is that most home pcs that use the 802.11 are behind a router/Nat which makes them useless for DDoS.

But using zombie hosts to launch a deauth ddos - well , the odds are almost close to zero.

And most ddos attacks usually use Syn or ack floods .. udp also sometimes.

Yea, and when people launch a ddos, they usually resolve you dns and hit any ip associated with it, coz they know once you get on an ip, your isp will null route that ip and will be willing to change your ip immediately.

One way to mess around with the skiddies is to point your dns to microsoft.com or cybercrime.gov when you are getting hit .. lol .. and it works a lot of times.

It is like re-routing their ddos to w/e ip you want to hit .


oh and yea .. if the owners of the WAPs or W gateways or w/e are using Wireless Intrusion Prev. Systems, (WIPS)s, then there is a good chance that they'll stop even the initial deauth attack.

holy shikeees .. google.com has a black back Tongue
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.5 | SMF © 2006-2008, Simple Machines LLC
Joomla Bridge by JoomlaHacks.com
Valid XHTML 1.0! Valid CSS!
Page created in 0.058 seconds with 22 queries.
 
Polls
Best for daily desktop use:
 
Support EH-Net
chicagocon2008f_125x200banner.jpg
ChicagoCon 2008f


Support EH-Net by
Buying all of your
Amazon items using
the search bar above.

cbtnuggets_logo_125.jpg
Try CBT Nuggets Free!
Recent Forum Topics
Vote For EH-Net

progenic.com
Click here to Vote!

Sadikhov.com
Top IT Cert Sites

binarica.com
Binarica Logo

Add to Technorati Favorites
technorati fave

chicagocon2008f_125x200banner.jpg
ChicagoCon 2008f
 
         
Advertisement

© 2008 The Ethical Hacker Network
Joomla! is Free Software released under the GNU/GPL License.