- tor-dev - lists.torproject.org

Checking Tonga's consensus (was: Consensus issues)
by Karsten Loesing 14 Oct '13

14 Oct '13

On 10/14/13 6:45 AM, metrics wrote: > WARNING: The consensuses published by the following directory authorities are more than one hour old and therefore not fresh anymore: Tonga Hi Damian, why do you check Tonga's consensus? Tonga is not a v3 directory authority but just a directory mirror. All the best, Karsten

2 1

More Hidden Services help needed: Guard enumeration
by George Kadianakis 14 Oct '13

14 Oct '13

Greetings, another important Hidden Service issue, is the guard enumeration attack that was described by the "Trawling for Tor Hidden Services: Detection, Measurement, Deanonymization" paper (in section VII) [0]. A trac ticket was created to fix this issue (#9001 [1]). The most popular solution so far seems to be the 'Virtual Circuit' concept. The idea here is that if you have already created a circuit to a Rendezvous Point, and then that circuit fails and you need to connect to a different RP, then maybe you could reuse the previous hops of the circuit and only change the last hop to the new RP. This is an interesting concept and useful in other occasions too. For example, if the network is flooded with CREATE cells (like during the recent botnet invasion) and relays reject them because of increased workload, then maybe clients shouldn't build entirely new circuits to send CREATE cells to different relays (because building new circuits increases the total load of the network even more). However, this concept must be designed carefully and in a security-conscious way. Reusing parts of old circuits to connect to new nodes might result in unexpected attacks. More solutions have been proposed in #9001, like "guard node layers", which need further investigations. If you think you can contribute to this topic, please write your ideas in this mailing list or in ticket #9001. Enjoy. [0]: www.ieee-security.org/TC/SP2013/papers/4977a080.pdf‎ [1]: https://trac.torproject.org/projects/tor/ticket/9001

1 0

Re: [tor-dev] [linux-elitists] Browser fingerprinting
by Eugen Leitl 14 Oct '13

14 Oct '13

----- Forwarded message from katana <katana(a)riseup.net> ----- Date: Mon, 14 Oct 2013 09:27:41 +0200 From: katana <katana(a)riseup.net> To: cypherpunks(a)cpunks.org Subject: Re: [linux-elitists] Browser fingerprinting Message-ID: <525B9CED.20907(a)riseup.net> User-Agent: Thunderbird Hi, > Check out firegloves. It's outdated, and I'd love to see it getting > some love, but it's a great POC for anti-fingerprinting in Firefox. In <http://www.cosic.esat.kuleuven.be/publications/article-2334.pdf> about their FPDetective Framework <http://homes.esat.kuleuven.be/~gacar/fpdetective/>, the authors wrote about Firegloves: "Additionally, Firegloves limits the number of fonts that a single browser tab can load and reports false dimension values for the offsetWidth and offsetHeight properties of HTML elements to evade JavaScript-based font detection. We evaluated the effectiveness of Firegloves’ as a countermeasure to fingerprinting, and discovered several shortcomings. For instance, instead of relying on offsetWidth and offsetHeight values, we could easily use the width and the height of the rectangle object returned by getBoundingClientRect method, which returns the text’s dimensions, even more precisely than the original methods. This enabled us to detect the same list of fonts as we would without the Firegloves extension installed. Surprisingly, our probe for fonts was not limited by the claimed cap on the number of fonts per tab. This might be due to a bug, or to changes in the Firefox extension system that have been introduced after FireGloves, which is not currently being maintained, was first developed. Although Firegloves spoofs the browser’s user-agent and platform to pretend to be a Mozilla Firefox version 6 running on a Windows operating system, the navigator.oscpu is left unmodified, revealing the true platform. Moreover, Firegloves did not remove any of the new methods intro- duced in later versions of Mozilla Firefox and available in the navigator object, such as navigator.mozCameras and navigator.doNotTrack." I add: OK, the naviagtor.oscpu issue can be fixed easily, but the timezone feature doesnt't work too with enabled JavaScript. --- Katana ----- End forwarded message ----- -- Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://ativel.com http://postbiota.org AC894EC5: 38A5 5F46 A4FF 59B8 336B 47EE F46E 3489 AC89 4EC5

2 1

Attention Otters
by Arlo Breault 14 Oct '13

14 Oct '13

Not sure the following is entirely clear or complete, but I tried to capture the concerns from the meeting and the ensuing discussion. Hope it helps. Arlo Attentive Otter Plan ==================== Goal ---- Add instant-messaging to the Tor browser bundle in order to provide a secure communication tool which supports the free flow of information online. Overview -------- Instantbird [1] is a cross-platform IM client based on Mozilla's XULRunner. The following presents the necessary steps to turn Instantbird into the future Tor Messenger. A Way Forward ------------- 1. Remove libpurple dependence This is a trivial amount of work and changes to the build to support it would be accepted upstream. They are already considering moving libpurple, and the added protocols it supports, to an add-on for reasons of licensing/code quality. JS implementations of the following protocols exist: XMPP, Google Talk, Facebook, IRC, Twitter, with Yahoo landing soon and AIM/ICQ started but further away. 2. OTR support Instantbird currently lacks support for OTR. Two pieces are needed here: a suitable OTR implementation, and an interface between the client and that library (essentially, the role that pidgin-otr plays). To get started, for the OTR library, a js-ctypes wrapper of libotr should be used in conjunction with the message observer API. Code [2] from a few years ago towards this end has been written but probably needs to be dusted off and extended. An effort is underway at Mozilla to implement OTR in JS using NSS, which could be dropped in as a replacement. A patch has been submitted [3] but it looks far from complete, so I wouldn't expect it anytime soon. When asked, they said it won't be ready for *a while*. Should the NSS implementation fail to materialize entirely, they would still be willing to take the ctypes wrapper and libotr, as it doesn't present any licensing issues. In his analysis, Mike suggested converting the ctypes wrapper to an XPCOM wrapper but it's unclear why that's preferable. The front-end side seems like a larger undertaking. This involves not only the interaction with the message observer API but handling the quirks in the various protocols (think /me in IRC), authentication including SMP, and importing and storing long-term keys. Sukhe estimated at least a month of development time and expressed an interest in being the one to undertake it. On the bright side, the Instantbird team seems eager for OTR support and this work will most likely be upstreamed. 3. Disable logging An add-on may be required to ensure certain desirable configurations, like logging disable by default. A difference in goals between UX for the average user and the TIMBB user may force us to maintain these changes. 4. Tor controller Tor Launcher will be used as the controller. Sukhe has already reported having this working. Using only JS protocol implementations means all traffic goes through nsIChannels, making proxy support fairly easy to verify. For DNS, network.proxy.socks_remote_dns should be set. DNS SRV should not be an issue seeing as how it isn't supported by Mozilla [4]. Should test for other UDP traffic leaks. 5. Messaging window Jail it to type=content. Preferably everything is displayed in plaintext, with HTML disabled or at least sanitized with an XSS filter [5]. Disable JS and other features. Make use of all the preferences from TorBirdy. 6. Installer and updates Leverage the work that's already being done on Mozilla's updater for the TBB. 7. Deterministic builds Deterministic builds for the TBB was a major undertaking. I can't imagine this case being any different, less the experience and groundwork already laid. 8. Sandboxing Come up with a practical, cross-platform way to sandbox the application. I don't have an answer here. Maybe you do. 9. Audit - Instantbird's render attack surface (content window, XSS filter, etc.) - Crypto in NSS and how JS uses it - Interface between the UI and OTR - Proxy by-pass - And more ... 10. Translations Instantbird is available in 14 languages, including French and Spanish. However, none are RTL and we want to support Arabic and Farsi. Messaging should already work for RTL languagues though, they've fixed a few bugs to ensure it, and reflecting the UI is reported to not be a ton of work. They are definitely willing to accept patches here. 11. Other considerations - Disable Instantbird's built-in auto-updater and crash reporter - For sure OTR on by default, but maybe disallow any non-OTR comm. entirely - CA verification: TOFU mode? Pin popular domains? - Disable older TLS/SSL suites - Consider the interaction between all three Tor bundles (FF, TB, IM). Tor Launcher could attempt to authenticate and read settings from an already running control port. - Choose a different default profile folder (to avoid picking up plugins and other unsafe settings) References ---------- [1] http://instantbird.com/ [2] https://gitorious.org/fireotr/fireotr [3] https://bugzilla.mozilla.org/show_bug.cgi?id=779052#c20 [4] https://bugzilla.mozilla.org/show_bug.cgi?id=14328 [5] https://mxr.mozilla.org/comm-beta/source/chat/modules/imContentSink.jsm

1 0

Chutney
by Charlie Belmer 11 Oct '13

11 Oct '13

Hey everyone, I am running some Chutney tests, and I want to go deeper than just running the bootstrap script and watching logs. Does anyone have advice on how to get the most out of Chutney? Thanks! Charlie

2 1

Pluggable transport biweekly meeting - Friday Oct 11
by vmonmoonshine＠gmail.com 11 Oct '13

11 Oct '13

Just a gentle reminder that the PT meeting is today in few hours: Friday - Oct 11 at CEST: 18:00 BST (Summer GMT): 17:00 UTC: 16:00 EST: 12:00 MNT: 10:00 PST: 9:00 @ #tor-dev on irc.oftc.net asn is going to moderate the session. (I have a very slow connection but I'll try to join) See you all there, Vmon

1 0

[Otter/Boisterous] WebChat report
by Colin C. 11 Oct '13

11 Oct '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 The Boisterous Otter project requires us to offer assistance to users through instant messaging. Current plan: use an XMPP server allowing anonymous connections from a web interface on one side, and have the support team use an authorized account for each language. The full report is viewable on the wiki at https://trac.torproject.org/projects/tor/wiki/org/sponsors/Otter/Boisterous…. (Thanks to Lunar & Erinn for assisting with this) - -- - -Phoul -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBCAAGBQJSV320AAoJEJ01Zvx7KO7YOgwQALmsJsnyVRUQ07CiqmxI+zle dNsROZvnNvhfPyza5McBjxckXQ5fD5tgCL5TMxzkBD3ICFQgZWByrX/bMN2ZFfx0 nh12TCs64TvMFAXnTVKRopBYfkas/uSs3PhM4vnPULJ+mDbFoMNKXkJP42s7jkGK D3q2+2k/yR9mu9IVLI9agKppR/L1Yq/lPzt9DFo0YZyIrI0nNrn9hFMCGQKuCl0u S40DWXXtUXn1a0qg4mFKGurrVua7ZvMeLqmZExhLMY8AnFBP40ELP0mmM0KkQg1F kW0ijmybEMcPwKsWltTNzwHj8pg7dLjD19/ZrVMjIhjinap0XDqcyjqJD3T1J4KE ti0+wZ7wXRfjJs6eA0otfu6xfF5JGYvOd9KniN567aPxjCVb2s1Gm60zA1PGQwAe npT3WTm4z02SHThEeS4NPJi1Vvt0WLGlHo6v5p0Fr+/g1dYqEmTK+6+kAHk+e2CS OXSvJ16rj7kncLMf9SImvbk0+nojC9qT/jgJQm4IBgxzSq+TVNwn463JpCdpCI6m wfqxAAqy3V+ZjG+KxZTrxRST/s5rdhKi+iiZUzgqx0SC/hEqWEoId+2q3DPyb69h exBV0eUE0XsHKi12qySOef7BfQXO2q15cdUcu+9VzqULcwP+xKd5NziVDMWd95fj skzK8xFSm4Epdpp65VWI =CISa -----END PGP SIGNATURE-----

1 0

[Otter/Cute] What's Cute in APAF
by Michele Orrù 10 Oct '13

10 Oct '13

Dear Team, For completeness' sake I am attaching to this email the report I wrote last week in order to summarize what the project APAF is about, and what there is in common between it and the Otter/Cute proposal. Eventually, feel free to add it to the trac page. After reading ["Cute" design and challenges], though, I think the report lacks an exhaustive description of APAF's threat model. In APAF's documentation, the only document vaguely resembling a threat model is embarassingly poor in terms of content[0]. If still possible, I would like to remedy writing a more detailed one during this week-end, maybe really pentesting the application with your help this time. Does this sound visible to you? ["Cute" design and challenges] <https://trac.torproject.org/projects/tor/attachment/wiki/org/sponsors/Otter…> [0] <http://apaf.readthedocs.org/en/latest/threat_model.html> -- mi.

2 1

Otter-related outreach, training & documentation meeting is at 1530h Pacific on Tuesday 2013-10-15
by Tom Lowenthal 10 Oct '13

10 Oct '13

It'll be in #tor-dev. If you want to be in the `to` field for future emails, announcements, and so on (rather than just reading them on `tor-dev` or `tor-assistants`), you should add your name, IRC nick, and email to the [Buoyant wiki page](https://trac.torproject.org/projects/tor/wiki/org/sponsors/Otter/Buoy…. -Tom On 8 October 2013 16:20, Tom Lowenthal <me(a)tomlowenthal.com> wrote: > People of Earth, heed my call! > > Based on feedback that Boisterous Otters has too much to do, and on > the fact that current Boisterous meetings are using their full time on > only half the topics, we're splitting Boisterous into two components. > > The first component is going to stick with the name Boisterous, and > deals with the first two activities: the helpdesk, and l10n. If you're > already working on these things, nothing is changing for you, and you > may be able to stop reading this right here. > > The second component is picking up the name **Buoyant**, and will deal > with the remaining activities: > * written materials on general communications security > * short videos multi-language training videos > * outreach to Iranians and the Iranian diaspora > * social-media, radio, and television outreach. > * training, and train-the-trainer work > > If that sounds like your cup of tea, kettle of fish, or flagon of ale, > then you should come help us get started by filling in this [Doodle][] > **by Wednesday afternoon** so that I can plan a meeting at the end of > this week or the beginning of next. Please fill out the Doodle if > you're definitely coming, and plan to be part of the core effort. If > you just want to show up to spectate or lend a hand, there's no RSVP > needed. > > I plan to look at the Doodle around 1700-1900h Pacific tomorrow, and > I'll use whatever it says then to try and pick the best time. I'll > send the time out to tor-dev, bcc-ing everyone who's expressed an > interest or answered the poll. That's 24 hours to give me your > availability, but only about 16 hours between when I send out the time > and the first possible meeting slot, so check your mail tomorrow > evening Pacific. > > If you have any specific suggestions about venue, format, or other > running *of the meeting*, and you plan to be there, please reply > publicly or privately before I send out the scheduling meeting. If you > have specific plans about how to git'er done, you can either email me, > or bcc me on a tor-dev-bound email with an appropriately-ottery > subject line. Do note, however, that other folks may not be able to > read your suggestions if you send them less than 12 hours before the > meeting. I'll try to read them anyway, but that's only out of the > kindness of my heart. > > Questions on a postcard please, > -Tom > > [Doodle]: http://doodle.com/pm5cfunbcwuyrwh6

1 0

Pushed HS proposals on git
by George Kadianakis 10 Oct '13

10 Oct '13

Hey Nick, I pushed the recent HS proposals to my torspec git repo. You can find them on branch 'hs_experimentation' at https://git.torproject.org/user/asn/torspec.git . My next step would be to write the anti-enumeration crypto part of xxx-hs-id-keys-and-onion-leaking.txt (as I started doing with c8b7e40). Feel free to work on them. Maybe send me an IRC message before you start touching them so that we don't do change the same areas of the spec.

1 0