| |||||||||||||
The dangers of weak random numbersThe dangers of weak random numbersPosted Feb 21, 2008 14:36 UTC (Thu) by kirkengaard (subscriber, #15022)Parent article: The dangers of weak random numbers
The more I learn, the more I find that it's dangerous to have non-cryptographically-strong RNGs. As you say, for anything more than picking which fortune entry to display, i.e. for anything greater than toy problems, you very quickly rise into areas where exploitation of predictability is problematic. People get used to practicing with a "sufficiently random" idea, and it leaks into implementations in areas where it becomes a problem. Communication is an access point; memory mapping is an access point; program input is an access point; data storage is an access point; data output is an access point. As to the OpenBSD response, this shouldn't be terribly surprising, given that having a name for security becomes momentum. We've seen them brush security problems under the table before, for love of keeping that "0 holes" statistic alive.
(Log in to post comments)
The dangers of weak random numbers Posted Feb 21, 2008 17:51 UTC (Thu) by bronson (subscriber, #4806) [Link] OK, what RNG would you choose and why? Please quantify. I'm just kidding -- a proper answer to this question would require writing a book. And endless debate. And the answer would change every five years. Ultimately, if you need soemthing done correctly to the last bit, would you really trust all the libcs in the world? My only problem with libc's rand() is the name. It's a flat out lie! No wonder it's confusing. It should have been called "prand" or "living_in_sin_rand" or something.
The dangers of weak random numbers Posted Feb 22, 2008 12:17 UTC (Fri) by jzbiciak (✭ supporter ✭, #5246) [Link] What I don't understand is why hardware RNGs aren't more common. A number of embedded CPUs contain them, but not the bulk of mainstream CPUs. They take up so little silicon compared to everything else, and provide such high quality results that they really ought to be ubiquitous. Sure, if you introduced them on mainstream CPUs today, it'd still be 5 years before they were everywhere, so you'd still have to implement other techniques in the meantime. But, those could be a bridge to a better overall solution rather than a road to more hand wringing in 5 or 10 years when somebody finds the next weakness.
The dangers of weak random numbers Posted Feb 22, 2008 2:19 UTC (Fri) by jschrod (subscriber, #1646) [Link] This also depends on the purpose of your RNG usage. Just this week, for example, I needed one to select a pseudo-random set of documents (out of 400,000) to pass them to human reviewers for inspection. For such a task, I don't care at all if it's cryptographically-strong or not, I simply use libc's rand(). I don't even care that it's very slow, getting the data for these 400,000 documents out of the database and generating/formatting them already needed enough time...
|
|||||||||||||