http://blog.gmane.org/gmane.comp.cryptography.viff.devel hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://comments.gmane.org/gmane.comp.cryptography.viff.devel/901 <pre>Hello, I am trying to test and use VIFF for the first time. I am new to python as well. I have followed the instructions given in the installation guide at viff.dk. The VIFF has been installed with the following versions: OS: Windows Vista Python: 2.5.4 Twisted: Twisted-11.0.0.winxp32-py2.5 OpenSSL: openssl-0.9.8g_win32 PyOpenSSL: pyOpenSSL-0.11.winxp32-py2.5 GMPY: gmpy-1.14.win32-py2.5 VIFF: 1.0 The path is set as: C:\Python25. VIFF is also installed in C:\Python25 \Lib\site-packages\ I have tried to sue both the test options given in the installation guide. But have the following error messages: ++method 1: python generate-config-files.py -n 3 -t 1 \ error: traceback(most recent call last): File "generate-config-files.py", line 57, in &lt;module&gt; from viff.config import generate_configs ImportError: No module named viff.config ++method 2: python millionaires.py --no-ssl player-1.ini (opening 3 seperate command prompts as instructed ) error: traceback(most recen</pre> Moha 2011-08-15T07:53:28 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/899 <pre>Are there any news on the Project? Be-D-O-SA documentation? Is it practical to use VIFF for a P2P application where peer loging off is unpredictable? Could their data be mirrored to another peer while keeping them secure? How does mirroring change the 'n/3' assumption? Just a few noob questions.. :°) I envision designing an economic system which will be based on having all the information. If I am correct MPC also protects against anyone playing agaist the rules. So MPC also makes such a design enforcable. Please continue updating this project and making it more practical. _______________________________________________ viff-devel mailing list (http://viff.dk/) viff-devel&lt; at &gt;lists.viff.dk http://lists.viff.dk/listinfo.cgi/viff-devel-viff.dk </pre> Apostolis Xekoukoulotakis 2011-07-14T22:40:38 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/893 <pre>Hello, My name is Abdelrahaman Aly, Im a Engineering PhD student at UCL, I was trying to install VIFF in CENTOS but I was not able to do it, there is any way to do it? Att. Abdelrahaman Aly </pre> Abdelrahaman Aly 2011-05-01T14:03:21 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/892 <pre>Hello, First of all, sorry for my english... I have problems when I use SSL. I explain: I have tried to run the file 'equality.py' in viff\apps. To do that, I generate config-files and certificates. Then I run 'equality.py player-X.ini -n 24' in three different shells. When I do 'equality.py player-1.ini -n 24' in the first shell, everything seems to be good. But when I write 'equality.py player-2.ini -n 24' in the second shell, python simply stops and I receive the following message: 'Python has stopped working. Windows is trying to find a solution to the problem.' I really don't understand what happens. I have exactly the same problem for all programs using SSL. However, everything works fine with no SSL. I use Windows 7 with: python 2.4.4 twisted 8.10 OpenSSL 1.0.0d (32 bit) pyOpenSSL 0.7 gmpy 1.10 Thank you for your help. </pre> Gilles Scuttenaire 2011-04-18T14:38:02 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/890 <pre>Hello, A real beginner question, I am afraid : I just started playing with VIFF. After installing the tools and running some of the apps example, I tried writing my own code, starting by modifying some of the examples. Taking shamir-share-open.py as a starting point, I started by doing some extra computations with the 3 shared values, e.g. computing a*c+b. Everything went fine. Then I wanted to compare some values. So I began by mixing Toft07Runtime with the default class, and adding a line like d = greater_than_equal(a,b) # (BTW, I know I could do the same with the operator &gt;=) Now the code stalls. Checking the file sort.py, the comparison result appears to be a deferred, so I tried adding a callback to d, but with no success. I also compared my code with compare.py, but I do not see where I deviate from what they do. Can anybody help me ? Sorry again if the question is stupid, I am a bit lost here, and searching mostly by mimicing other programs and trying whether it works isn't very eff</pre> Francois Koeune 2011-03-16T18:15:26 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/888 <pre> Good day! For a few years now we have been working on a practical way for creating applications that process confidential data. We are happy that we can now share some of the results with you. We have released the Sharemind toolkit for developing programs that can process confidential data separately from public data. In addition, confidential data can be processed using secure computation techniques like secure multiparty computation. You can find the software at http://research.cyber.ee/sharemind/ Currently, there is only a Microsoft Windows version, but expect support for more platforms in future releases. Please give these tools a try! There is a "Getting Started" tutorial on the Sharemind web page, that walks you through some basic features. We admit, that there are missing features and probably also some bugs. Please let us know of both! Our plan with Sharemind is twofold. First, we are looking for collaborators for building application prototypes or applications. We are already deplo</pre> Dan Bogdanov 2010-12-27T12:49:57 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/886 <pre>Here's a short note on how to update the viff.dk web page (from a private mail I got from Martin Geisler): The documentation found under http://viff.dk/doc/index.html is built from the doc/ directory in the main VIFF repository. As far as I remember, this is done by the buildbot, but it seems to be sick at the moment: http://buildbot.viff.dk/ That is, to update the viff web page, changes should be made to http://viff.dk/doc. The buildbot will then (when it's alive) update the viff.dk project and the web page. Regards, Thomas </pre> Thomas P Jakobsen 2010-12-15T14:32:35 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/874 <pre>Hello VIFF people, This place needs a bit of activity. First of all, I have mentioned that we can do linear programming in VIFF now. This is true. Together with Sebastiaan de Hoogh and Berry Schoenmakers I have tried making it actually viable in practice (non-toy examples work but take... a while :) ). It is much improved now but still quite slow. Turns out there was an algorithmic improvement to be made in the LP code itself. One thing that deserves to be mentioned is that memory usage here is a very real issue. But this got me thinking. What can be done better in VIFF? Apart from implementing more parts in C, I think we are looking for some fundemental changes in the communication model. In general, too much time is spent working on indivual shares, wrapping them in objects, sending them by themselves and so on. Also, of course, program counters account for much of the network traffic. We have already explored the easy improvements I think. We have put in a cheaper inner produc</pre> Mikkel Krøigård 2010-11-29T20:59:28 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/865 <pre>Hello, I posted a while ago that comparison doesn't work above a certain bitlength (969 bits to be exact). I've found out why. The following is called: rand.randint(0, (2**l) - 1) The value for l is somehow made into a float (in this case 1031.0). This works well for small values of l, but at some point it gives you an OverflowError that is hidden from the user. Just looks like the program is frozen. Shouldn't the error be shown? It works if you insert l = int(l) before the statement above. Obviously it needs to not be a float in the first place. I'd be happy to send in a patch for this if I could remember how that works. In any case, now you know. I can also say that I've tested it with 2000 bit numbers, and although it's not fast, at least now it works for huge numbers as well. Regards, Mikkel Krøigård </pre> Kroigaard, M. 2010-10-12T13:53:45 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/861 <pre>Hello, This is Kyung Hwang from Columbia University again. I have another question. Does Viff always need at least 3 participants? It seems to me it does. I modified "beginner.py" for two players because that file was simplest to modify, but when I ran the two players, I got the following errors: kwhwang&lt; at &gt;kwhwang-sim1:~/viff-1.0/apps$ python beginner2.py player-2.ini 20 --no-ssl Seeding random generator with random seed 3781 /home/kwhwang/opt/lib/python/viff/prss.py:43: DeprecationWarning: the sha module is deprecated; use the hashlib module instead import sha I am player 2 and will input 20 Not using SSL Listening on port 9002 #### Starting reactor ### ---------------------------------------------------------------- Program started Error: [Failure instance: Traceback: &lt;type 'exceptions.KeyError'&gt;: 3 /home/kwhwang/opt/lib/python/viff/runtime.py:317:stringReceived /home/kwhwang/opt/lib/python/viff/runtime.py:456:identify_peer /usr/lib/python2.6/dist-packages/twisted/internet/defer.py:280:callback /usr/l</pre> Kyung-Wook Hwang 2010-10-07T14:43:52 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/859 <pre>Hello, We've been trying to run an MPC LP solver (by Tomas Toft) for a while now and can't get it to work for examples of a more realistic size. One problem we're having right now is that using the Toft07 comparison, we can basically get VIFF to drop to 0% CPU usage and just sit there forever. What we did was run the "compare.py" test program from viff/apps but with Toft07 instead of Toft05, and with different bit lengths. Running it with anything up to and including 968 seems to work. 969 and beyond apparently makes VIFF freeze. If anyone has any ideas or can tell us what exactly we're doing wrong, it would be much appreciated. </pre> Mikkel Krøigård 2010-10-05T12:55:14 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/857 <pre>Hello, My name is Kyung Hwang, a Ph.D. candidate at Columbia University. I set up and tried to test VIFF's millionaires example. It works great when I ran it on a single linux machine. However, it does not work on three different machines. I am quite sure that all the three machines were correctly configured and they were actually connected to one another when I ran millionaires.py because they waited for all the three machines to start VIFF, and after they were all connected, each of them resumed its process and printed out its input information, for example, as follows (for player-1): "I am Millionaire 1 and I am worth 19 millions." However, that is it and it does not proceed any further, and it does not show the computation results which are supposed to be as follows: From poorest to richest: Millionaire 1 (19 millions) Millionaire 3 Millionaire 2 Synchronizing shutdown... done. Closing connections... done. Stopping reactor... done. It looked to me that each of those three players was trying </pre> Kyung-Wook Hwang 2010-09-27T01:20:02 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/852 <pre>Hi all, As you may or may not know, running the same VIFF program more than once using the same set of player configuration files is insecure when the runtime relies on pseudo-random secret sharing. This is the case for e.g. the PassiveRuntime. This is not a bug, but rather a consequence of the way in which we use pseudo-random secret sharing. In order to maintain security, a new set of player configuration files has to be generated for each run. In some practical settings this turns out to be inconvenient. So we recently added a changeset (1538:9d4f9551644c) that fixes this. It means that one no longer has to use fresh configuration files for each run. Instead, there's now a command line option called computation-id, e.g. python my_viff_program.py --computation-id=42 player-1.ini When using runtimes based on pseudo-random secret sharing, like PassiveRuntime, one can then safely reuse the configuration files as long as the computation ids are unique. That is, for each run with a given set of configu</pre> Thomas P Jakobsen 2010-08-12T10:04:01 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/850 <pre>Hi guys, I just want to let you know what I had a nice conversation today in the IRC channel for VIFF where there were 6 guys present (and one bot). They came from a organization that lobbies the EU for more openness: http://interfax.werebuild.eu/2010/05/31/werebuild-and-telecomix-cv/ It happens from time to time that people stop by and ask a question and I normally refer them to the mailinglist. But it would be great if a (current) VIFF develop would hang out in the channel from time to time. Don't worry about it being distracting -- most of the time the channel is very quiet. </pre> Martin Geisler 2010-08-09T09:36:12 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/847 <pre>Hi friends of VIFF, I've implemented Share and ShareList in C, based on a C implementation of Deferred. Using the C extension, benchmark.py and aes.py show a speed up between 50 and 100 percent. The code is in my repository: http://hg.viff.dk/mkeller To use the extension, first compile and install it: $ python setup.py install [--prefix PREFIX] Then, add the following lines two to your program before importing anything from viff or twisted: import viff.boost viff.boost.install(with_viff_reactor=&lt;True|False&gt;) If the parameter is True, the VIFF reactor is also installed, so this doesn't have to be done separately. There is a notable difference to the standard implementation of Deferred: Errbacks are disabled and exceptions raised in callbacks are not caught. This is for the following reasons: - The implementation was inefficient (too many isinstance() calls). - Errbacks are only used for error reporting in VIFF. However, I prefer VIFF to crash (and print a traceback) immediately if there is an exception</pre> Marcel Keller 2010-08-02T14:10:02 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/844 <pre>Hi 2 all! Good subscrubtion, thx! ________________________________________________ http://stydentkam.ru , http://ekzameny.net , http://referat-na-5.ru , http://hellocafe.ru , http://home-happy.ru , http://samkon.ru , http://scouts-russia.ru , http://bestxenon.ru , http://harmonia-plus.ru , http://kond-galereya.ru , http://divi-media.ru , http://v-peremen.ru , http://tonyfill.ru , http://v-peremen.ru , http://tbcom.ru , http://informkom.ru , http://aidsprint.ru , http://stydentam.ru , http://virtuallight.ru </pre> Мария Сергеева 2010-07-11T06:43:45 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/834 <pre>VIFF itself as well as most protocols implemented in VIFF uses the viff.util package for random number generation. This package in turn uses the random package in the Python standard library. This means that random numbers are generated using a Mersenne twister. As far as I can see, this is a problem, since Mersenne twister PRNGs are generally not suited for cryptographic usage. E.g. it is not known to pass the "next-bit test" and withstand the "state compromise extensions", see http://en.wikipedia.org/wiki/Cryptographically_secure_pseudorandom_number_generator. One solution would be to use the os.urandom() function instead. This has specifically been designed to produce cryptographically secure random numbers. (We should probably keep the old random generator, too. It is probably faster and not all random numbers used in VIFF and VIFF programs need to be cryptographically secure.) Let me know what you think about this. Kind regards, Thomas </pre> Thomas P Jakobsen 2010-07-06T13:22:58 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/833 <pre>Hello VIFF-team I would like to suggest the following patch to viff/comparison.py: 159c159 &lt; l = int(self.options.security_parameter + math.log(dst_field.modulus, 2)) --- &gt; l = self.options.security_parameter + math.log(dst_field.modulus, 2) otherwise the l in the next line: this_mask = rand.randint(0, (2**l) -1) is a float, and we get "34, Value out of Range" exceptions for big l. Best greetings Lars Krapf </pre> Lars Krapf 2010-07-05T10:39:52 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/832 <pre>Hi all, In the CACE project (http://www.cace-project.eu) we've just created a compiler that can translate MPC programs written for FairplayMP to VIFF programs. VIFF and Fairplay each have their benefits. If you write your MPC program in the Fairplay language, you can now freely decide which MPC engine you want to use to execute the program. Note that the compiler is not complete yet. There are still some performance issues and parts of the Fairplay language, SFDL, is not yet supported. The project is open source and you can find it at http://bitbucket.org/aicis/fairplay2viff Also, you can try out the compiler at http://smpc09.cs.au.dk:9091 For details on FairplayMP, see http://www.cs.huji.ac.il/project/Fairplay/fairplayMP.html Have a nice summer! Best regards, Thomas </pre> Thomas P Jakobsen 2010-07-06T11:21:20 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/829 <pre></pre> Martin Geisler 2010-05-17T22:55:08 http://comments.gmane.org/gmane.comp.cryptography.viff.devel/826 <pre>Hi, Is it still necessary to run `viff.reactor.install()` as described in &lt;http://www.mail-archive.com/viff-devel&lt; at &gt;viff.dk/msg00657.html&gt; in order to utilize the VIFF reactor? - If so, would it be possible to "fix" that? - If not, then the example apps would need to be updated accordingly. I will, by the way, be available in #viff &lt; at &gt; Freenode most of my time the next couple of days. Best regards, Joel Pettersson </pre> Joel Pettersson 2010-04-29T13:25:57 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.comp.cryptography.viff.devel