More Than One Third of Tutanota Emails Are End-To-End Encrypted
tutanota.com
tutanota.com
http://seclists.org/fulldisclosure/2015/Jun/58
How do they prevent man-in-the-middle attacks on their customers, even if their server is the one performing the attack?
"This is not a vulnerability in Tutanota. We have built Tutanota with multiple layers of protection for our users. We currently use TLS and DANE to protect authentication and data integrity and (only tunneled) RSA-OAEP and AES-CBC to provide confidentiality. We have always communicated this transparently, it is nothing new. Neither the confidentiality nor the integrity of our users' data has been at risk. However, we know that the implementation is not perfect regarding this detail. That is why we are going to implement the following features as soon as possible: - Signatures/MAC - 2-factor authentication - Algorithms resistant to attacks of quantum computers - Simple verification of downloaded Tutanota apps Regarding the described issue, we know of two possible attacks on AES-CBC. Neither of them is feasible against Tutanota users: - Bit flipping: You need access to the plain text email and you have to be the MITM. - -Plaintexts are available at the sender and recipient only. We use secure TLS algorithms and DANE to protect against MITM. - Padding oracle: There is no padding oracle in Tutanota. Tl;dr There is no known vulnerability in Tutanota. Security is the heart of Tutanota, and we will fix vulnerabilities immediately."
https://paragonie.com/blog/2015/05/using-encryption-and-auth...
And the answer is a mix of "we don't really understand crypto" and "you should trust us".
(Mix of random numbers and symbols)
I'd say avoid Tutanota.
> Padding oracle: There is no padding oracle in Tutanota.
Interesting claim. They're using PKCS5 which means that, yes, there is a padding oracle vulnerability, but triggering it is orders of magnitude more difficult than null padding.
https://github.com/tutao/tutanota/blob/tutanota-1.9.2+/nativ...
Why do you think that? Of course we (I am one of the founders of Tutanota) understand the difference of the two.
> They're using PKCS5 which means that, yes, there is a padding oracle vulnerability
Could you please describe the padding oracle? Knowing the padding algorithm is not enough to make up a padding oracle. The oracle must be able to decrypt the cipher text in order to tell the attacker if the padding is valid or not.
Because you put them on the same line item.
https://paragonie.com/blog/2015/08/you-wouldnt-base64-a-pass...
> Could you please describe the padding oracle? Knowing the padding algorithm is not enough to make up a padding oracle. The oracle must be able to decrypt the cipher text in order to tell the attacker if the padding is valid or not.
For fuck's sake, Google it. The top two search results explain it perfectly.
http://robertheaton.com/2013/07/29/padding-oracle-attack
https://blog.skullsecurity.org/2013/padding-oracle-attacks-i...
TL;DR if you aren't using authenticated encryption, because you're using CBC mode with a block cipher, an attacker can keep sending forged messages until it decrypts successfully. It's much easier to forge messages on zero byte padding than exploit PKCS(5|7), but attackers can still exploit padding errors to tell if their message decrypted.
> an attacker can keep sending forged messages until it decrypts successfully
There is simply no way for an attacker to find out if his forged messages decrypt successfully besides having full control of the recipients machine. The padding oracle is useless in this case.
Why is there so much resistance to implementing authenticated encryption? It would literally take 10 minutes to implement, with constant time MAC verification on the recipient's end. That includes the time to unit test it.
Hell, I could send a pull request right now to do it if your team doesn't have the expertise on board.
Better yet: use Libsodium.
https://paragonie.com/blog/2015/09/how-to-safely-implement-c...
We already stated that we are going to implement authenticated encryption. But when we do it we have to do it right. We have to keep everything backwards compatible and we can't enable one client to use the new implementation until all supported clients (JS, .NET (Outlook), Android and iOS) are upgraded. Implementing this kind of backward compatibility will not take only 10 minutes.
Please submit a pull request that keeps everything backwards compatible and works for all platforms. We are happy about everyone who wants to contribute.
Secure crypto is never backwards compatible with insecure crypto. If you want to expose your users to downgrade attacks, have fun.
Actually, there are good reasons not to: Spam filtering. People have forgotten how bad spam was. It isn't possible to fight spam centrally when all email is encrypted.
Pretending that there aren't any downsides isn't the way forward. It's important to be open about the issues at stake, then say what's possible with the tools that can be created.
Don't use passwords because you could forget them?
Better don't do backups at all, because a restore will most likely fail anyways.
What's next?
You could use something like zrtp once for fingerprint verificatiin.
Things like bitcoin have made it pretty easy to add digital money attachment to email as an anti-spam measure, first proposed as a proof of work system by Cryptographer Adam Back to introduce a cost to sending email, a bit like a computerised form of a captcha, something that doesn't affect normal users but is disastrously expensive for power users like spammers.
The combination ensures that 1) spam gets filtered 2) genuine emailers flagged as spammers get notified their email didn't arrive 3) important email can reach the recipient if it's worth enough (say $1, totally undoable for a spammer who'd end up with a cost-per-click on his email campaign of hundreds of dollars instead of cents, completely unworkable on any scale, but financially no problem for when say a stranger wants to send you a genuine email for the first time with a non-trusted key).
Not a bonus. Plenty of relevant emailing gets done 'cold'.
https://news.ycombinator.com/item?id=10210474
about the similarity between his ideas about Hitler's ideology and a short story Borges published in 1946. We had never communicated before; I heard of him for the first time through this press interview. He replied and said the connection was interesting.
I wouldn't be that opposed to ideas where people should put money at stake when making new unsolicited contacts, but I think a challenge is that people's wealth varies by many orders of magnitude, the recipient-perceived relevance of what they want to say varies by orders of magnitude, and the correlation between the two is often only moderate (maybe mediated by things like education, social capital, and fame).
There's also exactly zero overlap between encryption and proof-of-work. I don't see why you phrase it as if encryption makes proof-of-work either possible or easy.
One should seriously estimate and discuss the environmental implications of such a change before working towards it though, I think.
Lots of different approaches to encryption have pitched themselves as the easy-to-use replacement for PGP. Alas, so far, none of them have taken the world by storm.
That's good, though. PGP is the only one that's given true safety. Everything else has broken, either in theory or (mostly) in implementation. The problem is a social one, and it doesn't need to be solved quickly. It's fine if it takes several more decades before we find the right model.
Those that want to protect themselves can do so. Those that don't, don't. Not such a bad outcome, all things considered.
It seems like PGP's web of trust model was the mistake, anyway. There's not much wrong with the model of "here's my pubkey, and now you can send me data securely." The key can be intercepted or subverted, but what changed since the 80's was the rise of instant messaging. In 2015, you're often talking to people in realtime, and you want to talk and send data to them securely. To do that, you're exchanging keys in realtime, over some kind of chat program. (Even emailing someone a pubkey counts as a form of chat.) In that context, the web of trust is unnecessary.
You could also send with it detached key signatures from mutually trusted friends. This would add weight to your claim that this key belongs to you. Then, your recipient can check that your message's contents (including your pubkey) are signed by a key that your receiver can adequately measure authenticity. Does that seem right?
If a hostile MITM modifies your message and/or your key and signature, then these friend's signatures won't match. If they change these signatures as well, then those won't match the receiver's already trusted keys.
[Aside: One other way I have played with was to rate another signed message by a friend (who sent me his pubkey as an email attachment) with a likelihood that he wrote it, and then I checked if that message's signature matched. As a qualitative heuristic test, it did increase my confidence in that key significantly, though I would have liked to have some trusted 3rd parties witness his possession of it.]
How can someone gain this skill?
I am aware of only two general principles to do this: centralized certificate authorities (possibly chained), and decentralized assurance, a. k. a. web of trust. Their advantages and drawbacks seem pretty clear by now.
So, which is it, and if a CA (which I'd assume), who has the keys and how is crytographic trust obtained?