3 ms·
This isn't clear about "how" cracked it is. If it is going to take an attacker a year of CPU time, I'm not going to be too bothered. If it could conceivably run
by jstanley 13y ago
This isn't clear about "how" cracked it is. If it is going to take an attacker a year of CPU time, I'm not going to be too bothered. If it could conceivably run in a few minutes on a mobile phone then it's much more of a problem.
- gargarplex 13y ago(the mobile phone could talk to the cloud but a year is still expensive) i wonder how quickly you could solve that problem, cracking the wpa2 key, if you had $$$ resources to scale up. could it be usable? would make for a great android app :) i'd buy it!
- EddSeabrook 13y agoHave you read the actual paper or just the news article? I ask because I am thinking of buying the paper, but if it isn't any good then I won't.
- Jach 13y agoI've requested the paper through my school's inter-library loan, I'll know in about 4 hours if the librarian can get me a copy. If you send me an email and I get a copy I can forward it on to you. (Though it wouldn't surprise me if someone else 'liberates' the paper for everyone before then.)
- jgg 13y agoHowever, it is the de-authentication step in the wireless setup that represents a much more accessible entry point for an intruder with the appropriate hacking tools. As part of their purported security protocols routers using WPA2 must reconnect and re-authenticate devices periodically and share a new key each time. The team points out that the de-authentication step essentially leaves a backdoor unlocked albeit temporarily. Temporarily is long enough for a fast-wireless scanner and a determined intruder. They also point out that while restricting network access to specific devices with a given identifier, their media access control address (MAC address), these can be spoofed. That doesn't sound like something that would take a year of CPU time, but since no one seems to have actually read and analyzed the paper yet, who knows.
- krick 13y agoThat's right, not enough information. But, however your attitude might be the right one, I'm a little bothered already, because I didn't expect that to happen so soon. In fact, even the idea of "year of CPU time" doesn't sooth me much, because while most of your traffic might be "quickly-expiring" it's definitely not true for all of your traffic, and what's possible to do with a year of CPU time will be possible to do comparatively cheap very soon. And I guess we all remember stuff like Google collecting some wi-fi traffic anyway. So, it sounds somewhat scary to me. We don't even have anything better than WPA2 now, do we?
- fhars 13y agoAccording to some discussions on http://www.hackforums.net/printthread.php?tid=4129070 http://www.hackforums.net/printthread.php?tid=4129070 they "discovered" this attack: http://www.aircrack-ng.org/doku.php?id=deauthentication http://www.aircrack-ng.org/doku.php?id=deauthentication
- viraptor 13y agoJust to put this in some context, if we're talking about 1 core-year (I hope this unit becomes popular) what it really means is just over 500$ will give you the same result in one year / possible parallelisation. If the process ends with brute-forcing independent branches to look for an answer, that may be effectively an hour or less for each case. Not much of protection really...
- jstanley 13y agoThat's true, but it means someone has to have a $500 incentive. Nobody driving around looking to break into networks opportunistically is going to drop $500 per network.