News for hackers. Stuff that matters.
Re:Tor Browser and Tails Version Fingerprint by cloudjacker on Saturday May 28, 2016 @08:05PM (#11794155)
It detected Whonix as well

Not that it wasn't supposed to, just letting anyone know because its a waste of time for someone else to boot up their VMs just to see it, like I did.

Re:Tor Browser and Tails Version Fingerprint by programLyrique on Sunday May 29, 2016 @07:22AM (#11795958)
I am trying with Torbrowser 5.5.5 on MacOS X with javascript activated, and weak privacy settings, and the webpage detects neither that it is TorBrowser nor that it's on MacOS X.

I have also installed uBlock Origin; maybe it is what makes the difference?

Re:Tor Browser and Tails Version Fingerprint by ryan-c on Sunday May 29, 2016 @12:05PM (#11796822)
Installing any extensions in Torbrowser that it doesn't come with makes you very trackable.
Re:Tor Browser and Tails Version Fingerprint by programLyrique on Sunday May 29, 2016 @02:58PM (#11797694)
I'm quite sure I'm not the only one to use an ad-blocker with TorBrowser, so this would not be a unique fingerprint.
Re:Tor Browser and Tails Version Fingerprint by ecnahc515 on Sunday May 29, 2016 @03:35PM (#11797860)
It's not going to leave a unique finger print but not will certainly be able to track you...unless you completely trust your ad blocker
Re:Tor Browser and Tails Version Fingerprint by jgalt212 on Sunday May 29, 2016 @09:47AM (#11796344)
Idea: combine TOR + Guacamole.

Client runs guacamole client, one of the intermediate TOR nodes runs guacamole host. Fingerprinting takes place on intermediate node which is randomly chosen every time client starts up service. End result: Client remains anonymous to these sort of fingerprinting efforts

The downside is TOR, which is not terribly performant because of extra intermediate hops, will be degraded further by guacamole service.

Re:Tor Browser and Tails Version Fingerprint by noobermin on Saturday May 28, 2016 @07:04PM (#11793963)
Am I wrong in thinking that all this does is look at window.navigator? It doesn't seem to work for me though.
Re:Tor Browser and Tails Version Fingerprint by colejohnson66 on Saturday May 28, 2016 @07:13PM (#11793994)
Yes, you're wrong :) It actually opens up this page in the background:

    chrome://torbutton/locale/aboutTor.properties
If it exists, you're using Tor. Pretty obvious. For the version, it concats a few variables from `window.navigator', CRCs them, then compares the result against known CRCs. The code[0] is pretty easy to follow. I'm surprised Tor exposes these variables at all.

[0]: https://github.com/jonaslejon/tor-fingerprint/blob/master/to...

Re:Tor Browser and Tails Version Fingerprint by MichaelGG on Saturday May 28, 2016 @09:36PM (#11794387)
Isn't that only for Chrome? Tor Browser's based on Firefox eh. I thought XHR wasn't allowed cross-domain anyways, so shouldn't these things fail? Though I admit having no knowledge of how the addon system works.

Detecting Tor should be easier though, since exit nodes are published. So just check the IP?

I imagine all the variables have to be exposed because sites will break otherwise.

Re:Tor Browser and Tails Version Fingerprint by gruez on Saturday May 28, 2016 @10:47PM (#11794538)
>Isn't that only for Chrome? Tor Browser's based on Firefox eh

"chrome" in this context refers to the control elements of the browser

https://en.wikipedia.org/wiki/Graphical_user_interface#User_...

Re:Tor Browser and Tails Version Fingerprint by MichaelGG on Sunday May 29, 2016 @01:42PM (#11797306)
Right but I was unaware FF exposed a "chrome" Uri scheme.

I'm sure they have their reasons for letting pages make Xhr calls to add-ons like that but seems complicated.

Re:Tor Browser and Tails Version Fingerprint by developer2 on Sunday May 29, 2016 @04:47AM (#11795550)
I suppose Tails should at least update its support page regarding fingerprinting[1].

[1] https://tails.boum.org/support/known_issues/index.en.html#fi...

Re:Tor Browser and Tails Version Fingerprint by unsignedint on Saturday May 28, 2016 @05:48PM (#11793706)
Looks like it fails to identify Tor Browser 6.0a5-hardened...
Re:Tor Browser and Tails Version Fingerprint by r-w on Saturday May 28, 2016 @05:50PM (#11793715)
Cue them saying something along the lines of “our goal is to prevent tracing, not fingerprinting”. But I hope they’re hard at work on a fix regardless.
Re:Tor Browser and Tails Version Fingerprint by ryan-c on Saturday May 28, 2016 @05:54PM (#11793732)
Torbrowser includes patches to Firefox that make it more difficult to fingerprint individual machines, however this is at the expense of making it trivial to identify as Torbrowser when javascript is enabled. It doesn't appear possible to fix.
Re:Tor Browser and Tails Version Fingerprint by esnard on Saturday May 28, 2016 @09:51PM (#11794434)
Also, the requests seem to come from a node exit IP, which are public, so identifying a Tor user is easy.

Tor provides anonymity by giving all its users the same fingerprint. The fact that you're using Tor isn't a secret.

About the Tor version number in window.navigator, I guess they can't easily block it since the browser itself leaks some information via the features added in each release (e.g. a new JS API introduced in Firefox XX).

Re:Tor Browser and Tails Version Fingerprint by hackney on Sunday May 29, 2016 @03:24AM (#11795325)
Strangley, the tor site at one point mentions having javascript enabled by default. As far as I know, that is a no no for anonymous browsing using tor. But it also breaks most sites so there are not many times I actually use it. Even with javascript enabled, I'm sure you can still do a fair bit of anonymous browsing but I just don't trust java or the internet.