Why bother with the printer? :) There's a 99% chance (source: Tim's vague recollection of security stats 2007) that you can just plug in an access point to a network and go undiscovered for a very long time. If you wanted you could use a usb AP or just spoof the Mac if you wanted to go undiscovered for a little bit longer, or remain pretty invisible if you setup an AP policy to activate during only certain periods of the day to avoid supposedly regular admin wireless scans. There are plenty of ways to setup rogue APs and not be detected unless you have an always on Air*.* solution, EVEN if you have a written company security policy that forbids the use of wireless and you don't think you need to bother..... Wireless security is just as important to companies without wireless networks as it is to those with!
Rgds Tim Sent from my BlackBerry® wireless device -----Original Message----- From: "Eric Hacker" <[EMAIL PROTECTED]> Date: Wed, 21 Mar 2007 14:10:44 To:[email protected] Subject: Re: Wired detection of rogue access points Haven't we gone through this before? For each of you that thinks they have a way to detect a wireless access point using only the LAN, please demonstrate how you would detect this. A wireless router is hooked up to the network jack of a printer. The wireless router is configured to use the printer's MAC address. The wireless router is set up with the printer's new IP address as it's DMZ host[1]. From the outside, all port scans and probes are going to the printer. There might be some IP stack differences, but you'd have to have a very comprehensive database to figure that out, and the time to scan at that level could prevent that level of probing on large networks. From Mr. Waters, I expect no less than the results of an actual scan on a live network with this set up running on it. :) Now that was easy. No real expertise required on the person who set up the rogue access point, just a little cleverness. So lets say I want to put the rogue access point on your network. Same router, new firmware. My new OS is reconfigured a bit. The WAN port bridges to LAN1. WAN plugged in to wall, LAN1 plugged in to printer. All other ports and the wireless are configured for the private LAN on the router. My OS sniffs packets and determines the IP address in use by the printer. Now it statefully NAT's packets from it's private network to the printer's IP address. It filters return packets on the bridge so that the printer doesn't see any of the traffic. Now how do you find it over ethernet with scanning or probing? It doesn't respond to anything. It doesn't interfere with the printer's IP stack fingerprints when the printer is probed. Only watching the unusual traffic coming from the printer or scanning for the RF would pick this up. Oh yeah, heaven forbid that I go all out and not use normal wireless frequencies. Maybe pop in an EVDO card instead of an 802.11 one. Who would want their own Internet accessible back door into your intranet anyway? OK, so my OS isn't completely off the shelf, and I haven't had the time to sit down and make it work yet. The open source pieces are all there, however, just waiting for the right person to come along and duct tape them all together. Bottom line: Ethernet cannot be completely secured. Either encrypt everything, watch everything, or physically control access to everything. Regards, Eric Hacker, CISSP [1] I hate using the term DMZ for this use, but that's what is used on all the router configurations. ------------------------------------------------------------------------ Test Your IDS Is your IDS deployed correctly? Find out quickly and easily by testing it with real-world attacks from CORE IMPACT. Go to http://www.coresecurity.com/index.php5?module=Form&action=impact&campaign=intro_sfw to learn more. ------------------------------------------------------------------------
