Key Reinstallation Attacks

https://www.krackattacks.com/ "We discovered serious weaknesses in WPA2, a protocol that secures all modern protected Wi-Fi networks. An attacker within range of a victim can exploit these weaknesses using key reinstallation attacks (KRACKs). Concretely, attackers can use this novel attack technique to read information that was previously assumed to be safely encrypted.This can be abused to steal sensitive information such as credit card numbers, passwords, chat messages, emails, photos, and so on. The attack works against all modern protected Wi-Fi networks. " "The weaknesses are in the Wi-Fi standard itself, and not in individual products or implementations. Therefore, any correct implementation of WPA2 is likely affected. To prevent the attack, users must update affected products as soon as security updates become available. Note that if your device supports Wi-Fi, it is most likely affected. During our initial research, we discovered ourselves that Android, Linux, Apple, Windows, OpenBSD, MediaTek, Linksys, and others, are all affected by some variant of the attacks. " What are Lainchan's thoughts on this?

Oh soykaf, doublepost. Deleted + Moved my post to here. Key Reinstallation Attacks Breaking WPA2 >https://www.krackattacks.com >We discovered serious weaknesses in WPA2. An attacker within range of a victim can exploit these weaknesses using key reinstallation attacks (KRACKs). The attack works against all modern protected Wi-Fi networks. Depending on the network configuration, it is also possible to inject and manipulate data. >The weaknesses are in the Wi-Fi standard itself, and not in individual products or implementations. Therefore, any correct implementation of WPA2 is likely affected. To prevent the attack, users must update affected products as soon as security updates become available. We discovered ourselves that Android, Linux, Apple, Windows, OpenBSD, MediaTek, Linksys, and others, are all affected by some variant of the attacks. >Our attack is especially catastrophic against wpa_supplicant. Here, the client will install an all-zero encryption key instead of reinstalling the real key. (Because Linux + wpa_supplicant is the only implementation that strictly follows what the standard suggested, memset() the key to all-zero after installed. During the attack, it will reinstall the now-cleared encryption key, effectively installing an all-zero key) What if there are no security updates for my router? > Our main attack is against the 4-way handshake, and does not exploit access points, but instead targets clients. So it might be that your router does not require security updates. For ordinary home users, your priority should be updating clients such as laptops and smartphones. Like what I said in >>2852, WPA2 is not everything you need for serious security. Installing a secure VPN server on your router, e.g. OpenVPN is still the best way to guarantee Perfect Forward Secrecy. Now soykaf hits the fan, the advice proved to be useful.

fuarrrk. It's going to be a bad monday at the office today...

>>2941 Quite impressed, the white paper was really good and showed that we as world have put our faith in a protocol that was just accepted and not rigorously checked through and through. However it seems like the proper manufactures have been notified with enough time to release patches, however this is going to be a very painful day in the office tomorrow. reports, getting every system patched and arguing with people on why they should get it patched. fuarrrk me... Mikrotik is apparently unaffected for anyone wondering.

https://arstechnica.com/information-technology/2017/10/severe-flaw-in-wpa2-protocol-leaves-wi-fi-traffic-open-to-eavesdropping/ So apparently WPA2 is getting loads of vulnerabilities. And unlike the WPS vulnerability these seem to be in key parts of the protocol like handshaking ETC... Will this finally prompt people to start the standardisation of WPA3/4(suggesting that WPA3 might be just interim like WPA1 was)?

WPA3/4 is going to take a long time to instate especially considering that the general populace doesn't know how to do a firmware update. I'd estimate about 3 years for it being a manufacturing standard and 1 year for it to be slowly rolled out as firmware to the routers of companies who actually care.

>>2952 Well I don't care much about when the general populace will start using it(sounds like fun times to be honest). I do care when or if IEEE will make WPA3 so I can put it on my router.

I connect to free-wifi networks usually WEP with no actual security. I use the Tor browser liberally. Should I be worried?

>>2954 I connect to insecure wifi networks, change my MAC address, run a vpn and then use the Tor browser. when I'm done, I change my MAC address back. I feel pretty secure that way. Some people have said that running a VPN alongside Tor is a vulnerability though

Manufacture problem -> propose your own solution for said problem I guess it is being done so new protocols are rolled out and people who decide on them don't loose their jobs/positions.

>>2955 I recently started changing my mac address because I read it could be used to uniquely identify you. Didn't know that until then.

>>2954 Any non-encrypted traffic until now was at least encrypted on the network level with WPA2. Now it will be laid bare. In any case you should use WPA2 rather than WEP(can't hurt). Your security onion will be better if you use WPA2 even if its cracked right now.

>>2956 Gee what is your solution? Send everything in cleartext? Technology progresses and encryptions have weaknesses. As time moves on we find the weaknesses, and as technology has progressed we replace the old weakened encryption with new ones.

>>2958 >In any case you should use WPA2 rather than WEP(can't hurt). I think what he was saying is that he connected to WEP networks that he doden't own, though. The reason he uses WEP is because he can crack it and connect easily.

>>2960 Actually I just connect to open networks with wicd. They usually have a login portal or something. But can I enable encryption if I'm not the network owner? Can I still login?

>>2961 If you can get to the router settings sure you can. But doing so would kick out the other users wand would definitely get noticed.

Here's a good resource to check what companies and wifi providers have patched it and what ones haven't yet(spoiler, not many have): >https://char.gd/blog/2017/wifi-has-been-broken-heres-the-companies-that-have-already-fixed-it Imagine all the abandoned IoT devices that will be leaking personal information, basically indefinitely.

F​uck, how did this go undiscovered for so long? It's bizarre. And the fix for Android only appears to land on November 6th? Really, they know about it in advance but release the patch after the publication?

>>2961 100% of open public wifi networks are sniffed in some way.

>>2963 Maybe, but encryption can help on it greatly.

This was released very well. Everyone notified, major manufacturers with patches already out(but seriously, fuarrrk android, I'm not using WPA networks on my phone till I get that fuarrrkin patch). And yeah, everything about wifi is fuarrrked. This has probably been used in the wild for years. We need way better standards.

Getting MiTM on Wifi is trivial enough anyway. This new attack lets people who are not part of the network decrypt your traffic and in some cases inject things but it's not earth shattering. All Wifi networks you don't personally control should be considered malicious anyway.

>>2945 > Mikrotik is apparently unaffected for anyone wondering. they released a patch recently

>>2961 >Actually I just connect to open networks with wicd. They usually have a login portal or something. But can I enable encryption if I'm not the network owner? Can I still login? Irrelevant. The network owner can read all your traffic regardless of the Wifi encryption. Anyone else on the network can also MiTM your connection through ARP poisoning. You mitigate these existing problems and the new WPA2 attack the same way. But only using encrypted protocols like https or tunneling to a trusted machine with a VPN or SSH. >>2955 >Some people have said that running a VPN alongside Tor is a vulnerability though Both Tor and VPNs are encrypted so either will protect your traffic on insecure networks. >>2957 >I recently started changing my mac address because I read it could be used to uniquely identify you. If you are worried about that type of thing then look at a packet capture of your machine connecting to a network. Some DHCP clients will send things like your hostname and your exact kernel version number which can also be used to uniquely identify you. Mods: I think merging the /sec/ and /tech/ threads was a mistake...

Is key reinstallation attack the death of WPA/WPA2? What do you think lainons? https://www.krackattacks.com/

Supposedly it's patchable, as mentionned in the FAQ, so apparently not. Of course no one's going to know their APs need to be patched beyond security enthusiasts and some companies... who we all know are fairly (not) prompt to patch when needed.

No, just update your computer, and if you can, your APs. As long as your client is patched, you're fine. >Do we now need WPA3? >No, luckily implementations can be patched in a backwards-compatible manner. This means a patched client can still communicate with an unpatched access point (AP), and vice versa. In other words, a patched client or access point sends exactly the same handshake messages as before, and at exactly the same moment in time. However, the security updates will assure a key is only installed once, preventing our attack. So again, update all your devices once security updates are available. Finally, although an unpatched client can still connect to a patched AP, and vice versa, both the client and AP must be patched to defend against all attacks!

>>2976 it's mostly a client-side patch, and all major OSes already have it fixed.

There was a Lain ~a year ago (was it that long?) who made a thread about responsible disclosure. He claimed that he found a major flaw in the WPA protocol and nobody here believed him. Some people told him to do some less-than-legal things with the knowledge, but most told him to send Junk an email and get it in the Lainzine. I can't find the archived thread rn, but did Lain deliver? Was it our guy?

>>2980 Yeah, I remember that too. Too bad the boards got wiped because of mac user

>>2986 I don't know what your obsession with calling inexperienced people 'kiddo' is, but it's very easy to spot your writing style just about every time you act all condescending to someone else in the community. Maybe don't be such a prick? Just a suggestion. The world is already gross enough without lainions treating themselves like their mere presence is some kind of gift to the community. >>>Anyway, back on topic It's certainly a scary thing to think has been vulnerable for so long, but I'm super relieved that these major providers are largely on top of patching it before it becomes an issue. Props to the researchers for publishing it in such a way that we can learn from it, without just handing over a bunch of toys for script kiddies to abuse.

>>2991 >It's certainly a scary thing to think has been vulnerable for so long, If no one knew about it then I disagree. State intelligence might have known but they have better ways of capturing all your data. My biggest concern is the teenage script kiddie next door spying on my family's porn habits. Any serious attacker will setup a fake access point to get a robust man in the middle position on you so these WPA weaknesses don't bring anything new to the table.

Haven't seen anything on here about the KRACK vulnerability that was found a couple days ago. https://www.zdnet.com/article/wpa2-security-flaw-lets-hackers-attack-almost-any-wifi-device/ https://www.eff.org/deeplinks/2017/10/krack-vulnerability-what-you-need-know https://thenextweb.com/security/2017/10/17/krack-explained-like-youre-five-years-old/ List of devices getting patches: http://www.zdnet.com/article/here-is-every-patch-for-krack-wi-fi-attack-available-right-now/ It looks like https traffic is safe (for now), so using something like Https Everywhere should protect your data from this issue. It's a serious issue for older devices/devices from less popular vendors, as they might not ever be patched, but nothing apocalyptic. Thoughts?

>>3000 May be wrong there, but from what I understand https everywhere wouldn't help all that much. What it does is force pages that offer https versions to load that instead of http. It's a good function, and I use it. But there are still a lot of straight out http pages around, and if you access those, httpseverywhere won't be able to do soykaf. You just gotta be manually careful if you're on wifi and not access those at all. Also, there are tons of mixed content pages, with http linked content in https pages. Again, nothing httpseverywhere can do here. You need to straight out block that content. What I know of and use for this purpose is uMatrix but I don't think this option is on by default. As for patches personally I've dowloaded them for the Linux boxes, but I have an android device I need to use for work, and I haven't seen if it got any updated on that regard. Not a huge issue because I already treat it as compromised, since it belongs to Google not me, but this is a tad uglier than the usual datamining so I'd really like to know it's been fixed.