It looks like the security mechanisms commonly used by our web browsers are not done with getting bad press. The “Transport Layer Security (TLS)” technology and its predecessor, “Secure Sockets Layer (SSL)”, are cryptographic mechanisms designed to provide communications security over a computer network and are used literally everywhere. Recently, a group of cryptographers at INRIA, Microsoft Research and IMDEA discovered a new bug affecting this technology which is now known as the “FREAK” attack (Factoring RSA Export Keys). Devices shown to be affected by FREAK include iPhones, Android devices and Macs running OS X.
This vulnerability stems from something called the “Export-grade cipher suites” present in SSL/TLS that are still supported by many of our modern-day servers.
So what are export-grade ciphers?
Back in the 1990′s, when SSL cryptographic mechanisms were invented at Netscape Corporation, the US government implemented a strict “Export of cryptography from the United States” regime where , any cryptographic technology that would have to be exported outside the US should not exceed a certain strength (strength is defined by the algorithm and the key length used). The US did, of course, keep the stronger encryption techniques for themselves and these ciphers came to be known as “export-grade ciphers”.
How do FREAK attacks work?
For a client and a server to communicate over encrypted channels, both parties need carry out a handshake, negotiate and agree on the cipher suite (encryption mechanisms used both by the client and the server) to be used, and identify the best one which both parties can support. For a successful FREAK attack to happen, an adversary keeps sniffing the traffic (a so-called “man-in-the-middle” MiTM) between the client and the server.
When the negotiation is about to happen, the adversary sends a deceiving (spoofed) message to the server on behalf of the client requesting that “export grade” encryption to be used. The server agrees and responds with the weak export-grade set of ciphers (RSA) which the client accepts, resulting in a vulnerable encrypted communication being established.
The exploit has been described by the media as “potentially catastrophic” and claimed to be a side effect of “unintended consequences” of U.S. government efforts to control the spread of cryptographic technology.
The diagram below shows how the FREAk attack works.
What is striking is that a malicious “man-in-the-middle” could quite easily break the security of any website that allowed the use of weak export-grade keys. While the exploit was only discovered in 2015, its underlying vulnerabilities had been present for many years, dating back to the 1990s, and it is now very much in the public domain so shouldn’t be ignored
Web resources that are shown to be affected by the vulnerability include the U.S. Federal Government websites fbi.gov, whitehouse.gov and even nsa.gov. Based on geolocation analysis using IP2Location LITE, 35% of vulnerable servers were found to be located in the USA.
How can you Protect your data against a ‘FREAK’ attack
The only way to protect against the ‘FREAK’ attack, is to completely remove support for export grade ciphers from the server. By doing this,This will mean that even client requests for low strength cipher suites are responded to by the server with a failure to negotiate, stopping any attack in its tracks.
To find out more about how to stay resilient in cyberspace, contact us.