[NOTE: By the way, I am new... I posted a greeting in the new members thread...and I have been lirking here for awhile...and finally decided to become active because I really think the people here are more seriously focused on security and exploits (for whatever purposes)..
I am sure most of you heard of the talk at PasswordsCon in Las Vegas early August where it was announced that many more routers are vulnerable to a WPS exploit. Presented by 0xcite, a Swiss security company, it was announced that there were much more serious flaws than previously uncovered by the guy who wrote Reaver. The reason I bring this up, is because I am wondering if there has been an exploit written yet.
It is intriguing to me because: a) Dominque Bongard (the presenter) has stated that this only affected certain routers... but; b) but he only mentions one of them by name... Broadcom.
The fact that he mentions Broadcom, which is a
huge but does not mention the other vendor seems to be a conspicuous omission which may mean that this other vendor is ubqiquitous and the chipsets it produces with this flaw may be in countless routers (comprising numerous brands)—plus in the slides (below), <vendor2>, our mystery vendor, appears to have an even more severe flaw.
Essentially, two AES 128 hashes are computed and exchanged between the Registrar (AP) and the device (enrollee) in order to verify each to the to the other. The Registrar's primary flaw rests in its use of a pseudo random-number generator (PRNG) which is not random enough (not enough entropy). The implication is that these hashes therefore could be calculated offline in a matter of a few hours (if one can guess the PRNG state) and narrowed down to correct hashes which can be then used to falsely authenticate with the Registrar..
The mystery vendor sounds even worse... in that the random number generation schema is always null! That sounds like it could be extremely easy to hack.
I am not a crypto guy... but perhaps some people here are. Here are things I have gathered... I think there would be some notoriety if someone could figure an exploit out first... (unless someone knows of one already).
This was one of the American reports of the flaw in Engadget:
http://www.engadget.com/2014/08/31/wifi-protected-setup-flaw/and ArsTechnica:
http://arstechnica.com/security/2014/08/offline-attack-shows-wi-fi-routers-still-vulnerable/This is the Slideshare Presentation that was used at the talk.
http://www.slideshare.net/0xcite/offline-bruteforce-attack-on-wifi-protected-setupHere is the youtube presentation...It has limited access... you can only link to it if you go to the passwordscon website. One thing very suspicious I might add... is that the audio just gets dropped 3 minutes after Dominque Borgard starts his presentation. If you look in the comments to this video, someone from Reaver was complaining about the sound. Adrian Crenshaw, an organizer for PasswordsCon, had replied to the Reaver guy and others by saying that "Dominque (the speaker) walked too far away from his mic" Its obvious that this is complete bullshit. Adrian's comments about "Dominique walking away from the mic" have since been deleted as I am sure they seemed too inane and obviously not truthful.
https://www.youtube.com/watch?v=lD9E1bxGx9o&list=PLdIqs92nsIzRFk0OCN_uQiOkgtPiNk2mv&index=18This is a german article on the report of the exploty... I notice that germans like to discuss things on a much more technical level... People in the comments in this article also find that one vendor was not mentioned as very suspicious.
http://www.heise.de/netze/meldung/WLAN-Sicherheitsluecke-WPS-PIN-laesst-sich-manchmal-errechnen-2305568.html?from-mobi=1Below are older but much more technical pages (including a thread I found) on how WPS works in general... basically the vulnerability that Reaver exploited. I link them in case anyone who knows more about cryptography wants to dig into this a little deeper.
http://codeverge.com/grc.securitynow/wps-routers-brute-forcing-a-non-static-pin/1610685http://briolidz.wordpress.com/2012/01/10/wi-fi-protected-setup-wps/http://codeverge.com/grc.securitynow/wps-on-sn-337-4-digit-q/1610838
