Another high-profile vulnerability, nicknamed FREAK for "Factoring Attack on RSA-EXPORT Keys," is making splashy headlines. FREAK is related to a since-eliminated U.S. national security policy that required software makers use purposefully weakened encryption technology sold overseas.
Our SpiderLabs researchers have been closely studying the SSL threat since its existence went public earlier this week. We sat down with Karl Sigler, Trustwave threat intelligence manager, to learn what you need to know.
Q: There's another vulnerability with a catchy name. What is "Freak," and how serious is it?
A: It's one of those vulnerabilities that sits square in the "Medium" range. Although a successful FREAK attack would allow an attacker to decrypt all 'HTTPS' traffic between a victim and a vulnerable website, there are a couple of stumbling blocks for successful exploitation.
Q: Initially, researchers believed the potential number of affected people was limited - but it looks like that number has now massively grown. What happened?
A: At first it appeared that only Android, iOS and OS X devices were vulnerable to the encryption downgrade that FREAK performs, but researchers never sleep and it soon became apparent that Microsoft Windows was not immune. The addition of Windows systems to the list of vulnerable devices increased the attack surface greatly.
Q: So, at this point, should everyone assume their machines are vulnerable?
A: Pretty much. Since FREAK affects Windows and OS X computers, as well as Android and iOS tablets and phones, there are very few client devices that are not vulnerable. However, both the client (web browser), as well as the web server they are connecting to, need to be vulnerable. Web admins have been very busy in the recent days making sure that their systems are not vulnerable, and this limits the attack potential.
Q: What would be a typical way this vulnerability could be exploited?
A: An attacker would need to be on the same network as the victim. This would typically be a public network, like at a coffee shop or library. The attacker would then need to pick one encrypted web session where both the client and web server are vulnerable. They would hijack the session and force it to use weak encryption. They would then need to spend an estimated seven hours cracking the key. Once cracked, the key could potentially be reused against other sessions for a period of time, but that will differ from site to site.
Q: Have there been any active attacks and are we likely to see them?
A: There are no known attacks in the wild, and given the amount of work necessary to pull off FREAK, I would expect to see it used only in a very limited scope, if at all.
Q: What should businesses, depending on their size, do to protect themselves against FREAK?
A: Web server admins should look at their configuration and disable support for export level ciphers. For web clients/browsers, a website has been set up that can detect if you are vulnerable. You can test your browser here. As always, users should be especially cautious when using any public internet. You might want to wait to connect to your banking website until you are home.
Q: What else must be done to eliminate this bug?
A: Vendors like Microsoft, Google and Apple are in the process of creating patches to eliminate this vulnerability. Users should make sure to apply any patches as soon as they are made available.
UPDATE 3/10/15: Both Microsoft and Apple have now provided patches for the FREAK vulnerability. Microsoft on Tuesday patched all supported version of Windows with bulletin MS15-031. The Apple package, "Security Update 2015-002", patches OS X versions Mountain Lion, Mavericks and Yosemite. Apple also released a patch for iOS in iOS 8.2, available as an over-the-air (OTA) update. Google, meanwhile, announced that it has developed a patch and provided it to partners, but it's unclear when end users could expect to receive a patch.