- tor-dev - lists.torproject.org

Run With Limited Capabilities - GSOC
by Cristian-Matei Toader 01 Jul '13

01 Jul '13

Hello, My name is Cristian Toader, and I feel very excited about designing and implementing a capabilities based sandbox for the central Tor project, as part of the GSOC program. ---- About myself: I have been a Linux enthusiast for almost 6 years and have first started using Tor around 3 years ago. I am currently studying in the UK. In approximately one month I will be graduating the Computer Science programme at the University of Surrey, and plan on pursuing a master's degree in Advanced Computer Science at the University of Cambridge for the following academic year. I have completed a placement year at Qualcomm (UK) LTD which involved implementing and testing security solutions for the Linux Android OS. These were based on cryptography and the TrustZone run-mode of the ARM processors. Most of the development during the placement year was performed in C, with some tests written in Java and C++ for the upper layers. ---- About the project: The project I will be working on as part of GSOC is based on the "Run With Limited Capabilities" proposal [1] mentored by Nick Mathewson (nickm) and Andrea Shepard (athena). The project is still in the planning stage. I will start working on an appropriate design as soon as I finish my last exams, which is the 3rd of June. As part of the project I will need to: - investigate research papers regarding capability based sandboxes - get familiar with the Tor code structure - decide on whether there should be different states starting from which the tor program only has a limited set of capabilities, depending on what syscalls it should be able to perform; or maybe have a more complex approach based on a trusted process representing a root of trust (with no network interactions) which controls the capabilities of the processes it launches - integrate an appropriate solution - develop and run tests for the project A constraint agreed with nickm, would be that once the program capabilities are set they should not be modifiable (otherwise a potential attacker could have the option of first enabling capabilities and then execute privileged code). Some additional details can be found in tickets #7005 [2], #5219 [3], and #5220 [4]. I will try to keep everyone updated. I am looking forward to advice and suggestions. If anyone needs to contact me, this is my primary email, my irc.oftc.net username is ctoader, and I am geographically located in GMT+2. Best regards, Cristian Toader. [1] https://www.torproject.org/getinvolved/volunteer.html.en#limitCapabilities [2] https://trac.torproject.org/projects/tor/ticket/7005 [3] https://trac.torproject.org/projects/tor/ticket/5219 [4] https://trac.torproject.org/projects/tor/ticket/5220

5 6

[GSOC] Status report - Tor capabilities
by Cristian-Matei Toader 01 Jul '13

01 Jul '13

Hello tor-dev, As a small reminder the purpose of the project is to create capabilities based sandboxing for Tor, which may only allow the program to execute a number of predefined syscalls. For the past 2 weeks: - I have consulted with Nick Mathewson (nickm) and agreed upon using seccomp2 [1], and more recently a library built on top of that called libseccomp [2]. - I have set up a public remote branch [3]. - We have agreed on a 3 step plan for the project: 1. General sandbox based on a single (permisive) filter which restricts tor to using a number of syscalls. 2. Add configuration option for step 1, if any parts were broken in phase 1 by adding capabilities, they can be re-enabled at the cost of security. 3. Figure out what functionality should be split into separate processes, based on our experience from step 1 and step 2. - So far I have implemented step 1 using both libseccomp and seccomp2 [3]. Step 1 was developed in such a way that nothing from tor should be broken at the moment; What this means is that sandboxing currently exists in the remote branch, but is fairly coarse and will need some fine tuning at a later stage such as only allowing specific files to be open, or allowing the exec syscall to be called with specific parameters. These days I will be adding command line support, which is basically step 2, which will be followed by a code review and merge in the master branch. [1] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-precise.git;a=blob;f=Documenta… [2] http://sourceforge.net/projects/libseccomp/ [3] https://github.com/cristiantoader/tor-gsoc-capabilities

2 2

New flash proxy facilitator domain fp-facilitator.org
by David Fifield 30 Jun '13

30 Jun '13

I moved the flash proxy facilitator to a new domain, fp-facilitator.org. This is to get it away from bamsoftware.com, which also has a lot of unrelated stuff. The old facilitator name tor-facilitator.bamsoftware.com will continue to work (the DNS for both points to the same place). https://trac.torproject.org/projects/tor/ticket/7160 is the ticket. David Fifield

2 3

Status report - HTTP pluggable transport
by Chang Lan 30 Jun '13

30 Jun '13

Hi there, During the first two weeks of my GSoC project, I have implemented a HTTP CONNECT-based pluggable transport. In short, I use HTTP CONNECT semantics to establish a secure channel between the client and the bridge. Specifically, this is the scenario: 1. Connection establishment: +------------------------------+ |CONNECT bridge_ip:443 HTTP/1.0| |User-agent: blablabla | +------------------------------+ | | +----------+ | +----------+ +----------+ | client |----->| proxy |----->| bridge | +----------+ +----------+ | +----------+ | | +------------------------------+ | (Establish a connection) | +------------------------------+ 2. Data relay +------------------------------+ | (Encrypted Payload) | +------------------------------+ | | +----------+ | +----------+ +----------+ | client |<----->| proxy |<----->| bridge | +----------+ +----------+ | +----------+ | | +------------------------------+ | (Encrypted Payload) | +------------------------------+ I hope the diagrams above are self-explanatory. It is only my initial attempt to get familiar with HTTP protocol. Once I make sure it is working correctly under squid proxy I will upload it to the repository. The use of CONNECT method is restricted in many networks, so it is better to implement the HTTP transport using the usual HTTP methods as POST, GET, etc. In the next stage, I plan to implement a new HTTP transport based on BOSH[1]. There are many issues to care about (in order of priority): * bi-directional data transfer over HTTP * proxy cache * HTTP request/response encoding * encryption * scanning resistance Have a nice weekend! [1]: http://xmpp.org/extensions/xep-0124.html Best, Chang Sent from my mobile device. Sorry for the brevity.

3 2

Re: [tor-dev] Gitian builds in VirtualBox with Vagrant
by Arlo Breault 29 Jun '13

29 Jun '13

> This is rather exciting. Do you think that this method can be adapted to > build the pluggable transports bundles? Yes. I originally started working on this so I could submit a patch to include the tor-fw-helper in the TBB. But since the helper is only needed in the PT bundles, reworking their build process seems the more productive route. I'll see what I can do. > This is currently done > semi-manually, with a makefile that builds the pluggable transports, > unzips the bundle, copies in some files, and re-zips the bundle. > > https://gitweb.torproject.org/pluggable-transports/bundle.git/blob/HEAD:/Ma… > > We also have instructions for setting up a build VM from scratch--which > it sounds like is what Vagrant does--but I guess we would instead use > whatever is the build machine for the vanilla bundle. > > https://gitweb.torproject.org/pluggable-transports/bundle.git/blob/HEAD:/Ma… > > It would solve a lot of problems for us to have the PT bundles built at > the same place and time as the vanilla bundles.

3 2

Vidalia 2.0 - an complete rewrite
by Leo Unglaub 28 Jun '13

28 Jun '13

Hey friends, i want to inform you that i am working on a complete new control tool for Tor. I know some people use Vidalia for it, but i never liked the program and the people i showed vidalia often got confused. So i started to rewrite it from scratch and i am close to the release of the source code. (Currently i need to clean up some classes in there) I had a few things on my feature list for the new control tool and i want to share/discuss it with you. Because maybe some of my ideas are bad or maybe you have some additional ideas for features. A: "native work flow". Thats one of the main goals of my rewrite. The current Vidalia does things a little bit different than people are used to do it. So they get confused, because it does not behave "naturally". So instead of QT i used GTK and followed as always the GNOME HIG. (http://developer.gnome.org/hig-book/3.5/) B: "Multi configuration setup". What do i mean by that? Well, in my case for example i use Tor on my laptop and i am often in different places where i need different settings for tor. For example, if i am at home i can give tor all the bandwith i have, because at home i don't care. In my company on the other hand i cant do that. I am allowed to run Tor in the company, but with a bandwith limit of 2000kbit/s. (Yes, my company allows this ... well, i am the technical directory, so it was my call :) ) But i need a different Tor configuration there. Or if i am giving a talk and only have mobile internet connections i need every bit of bandwith for checking my emails so i want no server, only Tor client support. Currently i have multiple torrc files for that and always move them arround. That works for command line junkies like me, but even i am anoyed to do this multiple times a day. So i added the "multi configuration" support into the tool. This allows you to define multiple configurations and simply change them with a click from the gnome2 notification panel. C: "GUI for config settings". Well, some users are simply not ready to modify config files on there own. So i provide a simple gui (tab based) for all config options i found in the normal torrc file. Always with an explonation (with needs spellchecking, because my spelling sucks in german and is even worse in english :) ) so the user can deside what an option would to. D: "Notifications": My control tool uses the normal desktop notifications to inform the user about changes in the running tor application. For example if there is a problem building the circuit or something you get the notifyd message and can simply react to events without having the arm log open all the time. Thats just a quick overview of what i am duing right now. The source code will be released soon and is of course 100% open source. If you have some addiional ideas, feedback, wishes, ... i would like to hear them. Thanks and greetings Leo PS: I am Germam and very sorry for every spelling error i did above. -- Leo Unglaub Nachreihengasse 39 / Top 7 A-1170 Vienna Austria Phone: 0650 / 6575103 Website: http://www.leo-unglaub.net Twitter: http://twitter.com/LeoUnglaub Send with Mozilla Thunderbird on Linux (Debian)

9 24

Status Report - Steganography browser addon a GSOC 2013 project
by Hareesan 27 Jun '13

27 Jun '13

Hi all, In the first 4 weeks of my GSOC project steganography browser add-on, I have implemented the basic UI parts in first two weeks, and later I spent more time on web content downloading part. Up to now, I have finished downloading image and get them as bytes. Now when you right click on an image and click on test button, you will be getting the byte format of image which can be used with steganography algorithms to write messages. Since I'm mainly concentrating on features rather than algorithms now, my plan for next 2 week is to extend the file downloading for video and sound files. You can find the source code under git repository [1]. [1] https://github.com/rharishan/Steganography-Browser/ -- Hareesan

1 0

Re: [tor-dev] Torsocks development status
by David Goulet 27 Jun '13

27 Jun '13

Ian Goldberg: > On Wed, Jun 26, 2013 at 03:55:58PM -0400, David Goulet wrote: >> Hi everyone, >> >> For those who don't know, I've been working on a new version of Torsocks >> in the last three weeks or so. >> >> https://lists.torproject.org/pipermail/tor-dev/2013-June/004959.html >> >> I just wanted to give a quick status report on the state of the development. >> >> The DNS resolution is working for domain name (PTR) and IPv4 address. >> Currently, Tor does not support IPv6 resolution but the torsocks code >> support it. >> >> Hidden service onion address resolution is also working using a "dead IP >> range" acting as cookie that is sent back to the user and mapped to the >> .onion address on the hijacked connect(). >> >> I've changed quite a bit the configuration file (torsocks.conf) to fit >> the style of tor (torrc). At this point, the tor address and port can be >> configured as well as the "dead IP range" mention above. More is coming >> but pretty simple for now. >> >> Logging is working, connection registry and thread safety as well. There >> is also a compat layer for mutexes and once I start porting the project >> to other *nix system (BSD, OS X, ...) probably more subsystem will be >> added to that compat layer. >> >> So, in a nutshell, some libc calls still need to be implemented, *moar* >> tests and other OS supports. I'm confident to have a beta version to >> present to the community in a couple of weeks (if nothing goes wrong). >> >> Feel free to browse the code, comment on it, contribute!, etc... >> >> https://github.com/dgoulet/torsocks/tree/rewrite > > Are non-blocking sockets, select/poll/etc. (especially at connect() > time), and optimistic data on the to-do list? Yes! Good point I should have put the todo list. So yes, non block socket support. For optimistic data, it is kind of tricky. I can use it for DNS resolution without a problem because torsocks control the complete flow of data from opening a SOCKS5 connection to closing it after the DNS response is received however for actual real data (sendmsg, send, ...) a connect is needed before so it would means that a connect() call will return "yes OK socket connected" but where in fact it is not really true. So, when the first data are sent, there is a possibility that the Tor connections failed or even we block for an unknown amount of time during the send*/write() call. Now the question is, is this the kind of behavior that would be acceptable meaning basically lying to the caller at connect() and possibly blocking I/O calls and returning something like ECONNRESET or ENOTCONN if the Tor socks5 connection fails. This is *real* tricky especially with non blocking socket, if torsocks needs to do some possible blocking call for the SOCKS5 replies during an I/O call from the caller that is not suppose to block. Furthermore, having pending data that *might* come at any time on the connection from the SOCKS5 negotiation, the caller could put the file descriptor in poll() mode, wake up and try to receive the data but where in fact it's the socks5 reply... it's possible to handle that but it seems here a VERY intrusive behavior. Does optimistic data worth it here vis-a-vis the complexity of handling that it and high intrusiveness ? Cheers! David > > Thanks, > > - Ian

3 4

Status Report - Evil Genius
by Johannes Fürmann 27 Jun '13

27 Jun '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi *, The first two weeks of working on my GSoC project[0], which has been dubbed "EvilGenius", are nearly over. and I just wanted to share a bit of the experience I made during that time. EvilGenius is a tool that - much like Descartes' idea of an "evil genius", it "presents a complete illusion of an external world, including other minds, to Descartes' senses, where in fact there is no such external world in existence." (Source: Wikipedia), it distorts the guests' view of the real internet, in short, it creates a virtual environment that simulates censorship in order to verify OONI's nettest modules. So, after getting started with vagrant, the virtualization abstraction layer used for my project, i started simulating a virtual netwotrk and configuring the hosts to act as if they were connected to each other and the rest of the internet through a router that is part of the simulated environment. the "precise32" box is used as base image for all nodes, and I created scripts that configure the network and packed it up in a nice vagrantfile. Then I began deploying ooniprobe and oonibackend and executed the first successful tests. Next week, I will begin working on censoring environments. Have a nice weekend everyone! Johannes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.20 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQIcBAEBAgAGBQJRzKDWAAoJEAgie5o3Q/JpUhAQAIt2WEpDsNyLVDuSdMBM0ETf q0sOxdx20Q6o77J93FJ4ZBJRzpBX4mlsE4t22Dl/TPgnHu3b+DG+dvVhfCBrQOqw wGUuA6/X/khCikAOo+f3T9fNtaHKw54phW1Tw9Yj3Gea7SBN5ZikfXKjIIHOT+lu n0xqWVzxAxqGOETy0byDznsdgc2OfaR2NkdxzCsLlZTLtE75h/EQ3U96ZCej8lw/ xmZZUD+q+4ZCE6bBKTebufQeh0eI/rUPgo0UX1rNLgUapfCs9oLVIF+8nwBKBU0j E2myEOUbXMCxtTURbHikRQKzGKlsXMCEqqSatik37Rc3VaeVdkxNiiyXFVRwzTr0 IwRTibzJ8hD+aqh9d9VDD6s7dIdRio42p+XzD8jswFltEyYcbBGeAFFiybdNeeE2 EFb7n7iHz7Gr1csezyma1fdqvvMHZaaeCMTqeRENQtBaazOnqDwrGJn83r/FC7QF m8i+3eVDm8suM7CzYvuYi8YSIhF7yMVerIBOcBSZ2A929SVKtQFT4by1V2Jc4pz+ 0ZJV4teS8xct0wjhpDLANBCXgnV07OO6Z6EgLhcWpJBWoZDONDaxikH7clmTlzdg w/9ZktL3Ac3PjhSdMjOI/TKqc68+CHQ7ln1mG+YtSimMjf5Abpx1wPgLLILmzFNS lpZ2vev2pW0DsCnvs946 =abNV -----END PGP SIGNATURE-----

1 0

Gitian builds in VirtualBox with Vagrant
by Arlo Breault 27 Jun '13

27 Jun '13

If anyone has VirtualBox and Vagrant installed, https://www.virtualbox.org/ http://www.vagrantup.com/ and wants to try the new TBB deterministic builds, this worked for me, > git clone -b vagrant https://github.com/arlolra/tor-browser-bundle.git > cd tor-browser-bundle/vagrant > make Mike, there's a few patches in there you may want, https://github.com/arlolra/tor-browser-bundle/compare/vagrant Or you can take the whole thing. Enjoy, Arlo

3 2