A critical iOS vulnerability that Apple patched on Friday gives attackers an easy way to surreptitiously circumvent the most widely used technology for preventing eavesdropping on the Internet. That made the security bug about as dire as one can be. Now, there's strong evidence that the same flaw also exposes sensitive e-mail and Web communications on fully patched versions of OS X, with no indication that there is a patch currently available for the millions of people who use the Mac operating system.
At this early stage, the vulnerability has been confirmed in iOS versions 6.1.5, 7.0.4, and 7.0.5, and OS X 10.9.0 and 10.9.1, meaning it has silently exposed the sensitive communications of millions of people for weeks or months. Security researchers haven't ruled out the possibility that earlier versions are also affected. Readers should immediately update their iPhones and iPads to versions 7.0.6 or 6.1.6, preferably using a non-public network.
For the time being, people using Macs should avoid using public networks, a step that can thwart many criminal eavesdroppers but will do little to prevent surveillance by the National Security Agency and other state-sponsored spies. Because the Google Chrome and Mozilla Firefox browsers appear to be unaffected by the flaw, people should also consider using those browsers when possible, although they shouldn't be considered a panacea.
The flaw, according to researchers, causes most iOS and Mac applications to skip a crucial verification check that's supposed to happen when many transport layer security (TLS) and secure sockets layer (SSL) connections are being negotiated. Specifically, affected apps fail to check that the ephemeral public key presented by servers offering Diffie Hellman-supported encryption is actually signed by the site's private key. Attackers with the ability to monitor the connection between the end-user and the server can exploit this failure to completely decrypt and manipulate the traffic by presenting the app with a counterfeit key.
An attacker "can basically set up a connection and pretend to be Google.com," Matt Green, a Johns Hopkins University professor specializing in encryption, told Ars. The attacker "can basically say: 'Hey I'm Google, here's my signature. And since nobody is actually going to check the signature, [the attacker] just puts nonsense in there."
.
.
.
.
.
.
Testing 1, 2, 3
Langley has set up this test page, which is signed with a key that doesn't match the underlying TLS certificate. Apps that are able to access the text without generating an error are presumed vulnerable. While Chrome is one of the few applications that isn't susceptible to the attack, Langley said it wasn't clear if the update mechanism for the Google browser is vulnerable. If so, it would mean that attackers may still be able to compromise Chrome users, at least in some cases. It's unclear if Firefox is vulnerable to similar techniques, but until Mozilla representatives weigh in, readers should leave open that possibility as well.
Full Article
