Next question: can the NSA crack Tor keys?


( —"After more revelations, and expert analysis, we still aren't precisely sure what crypto the NSA can break. But everyone seems to agree that if anything, the NSA can break 1024 RSA/DH [DH refers to Diffie-Hellman] keys." With that Friday blog post, Errata Security CEO Robert Graham ignited a spark of outside posts reporting on Graham's observations about Tor.

"The problem with Tor is that it still uses these 1024 bit keys for much of its crypto," he said, "particularly because most people are still using older versions of the software. The older 2.3 versions of Tor uses keys the NSA can crack, but few have upgraded to the newer 2.4 version with better keys." Tor is short for The Onion Router, a service that offers anonymous browsing. Tor takes a user's Internet traffic and masks its location. He suggested that the Tor community do a better job getting people to upgrade to 2.4.

His speculation follows an exercise where he ran his own "hostile" exit node on Tor and surveyed encryption algorithms established by incoming connections. TOR still uses 1024 RSA/DH keys for much of its crypto.

About 76 percent of the 22,920 connections that he polled used some form of the older version keys. "Only about 24 percent of incoming connections were using the newer software," he said.

With the newer keys. the operations involved are more computationally intensive. According to the NSA site posting back in 2009, the US National Institute for Standards and Technology recommended that these 1024-bit systems be upgraded to something providing more security. The NSA discussion, titled "The Case for Elliptic Curve Cryptography." said that the US National Institute for Standards and Technology recommended that these 1024-bit systems were sufficient for use until 2010. "The question is what should these systems be changed to? One option is to simply increase the public key parameter size to a level appropriate for another decade of use. Another option is to take advantage of the past 30 years of public key research and analysis and move from first generation algorithms and on to elliptic curves."

"Of course, this is still just guessing about the NSA's capabilities," noted Graham.

Explore further

Protection for whistleblowers: Computer science professor working on system that would allow for secret data transfer

More information: … le.html#.UiyNAca1HA5 … nsa-researcher-says/

© 2013

Citation: Next question: can the NSA crack Tor keys? (2013, September 9) retrieved 16 October 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Sep 09, 2013
Doesn't it seem to hint that if the NSA is now recommending a new encryption standard that it is generally safe - but they already know how to undo it?

Sep 09, 2013
verkie; they are referring to using 2048 bit keys instead of 1024 bit keys. some people who believe the performance cost is worth the extra security use 3072 bit and even 4096 bit keys.

The stronger the key, the longer it takes to crack.

you might want to learn more about how this stuff works:

Sep 09, 2013
In November 2007, Bruce Schneier wrote in Wired: "there's an even bigger stink brewing around Dual_EC_DRBG ... the algorithm contains a weakness that can only be described a backdoor" placed there by NSA.

By no stretch of the imagination am I a cybersecurity or cryptography expert, but I'd be very very leery of using elliptic curve for anything at this point.

Sep 09, 2013
I'm planning to make a bootable DVD with SSH and a simple one-time pad program with a hashing function, and a lot of quantum-random data for the pad. Nothing can break that, but I will have to snail-mail or hand-deliver copies of it.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more