| |||||||||||||
/dev/urandom/dev/urandomPosted May 11, 2006 7:37 UTC (Thu) by ekj (guest, #1524)Parent article: On the safety of Linux random numbers Actually, I'm fairly certain urandom is *not* purely algoritmic "pseudo-random" numbers. My impression was that urandom would give you real random numbers based on the collected entropy-pool, but that it (and this is the difference to /dev/random) will *not* block when entropy gets "too low", but instead continue with pseudo-random numbers. Did I misunderstand this ? Or did our editor misunderstand it ?
(Log in to post comments)
/dev/urandom Posted May 11, 2006 9:38 UTC (Thu) by kleptog (subscriber, #1183) [Link] Your understanding matches mine. I've seen reports about people complaining that programs consuming large amount of data from /dev/urandom cause programs using small amounts from /dev/random to block. On other systems the data comes from seperate pools whereas on Linux you get the same data from both, except /dev/urandom doesn't block but becomes PRNG when the pool is empty.
/dev/urandom Posted May 11, 2006 20:35 UTC (Thu) by copsewood (subscriber, #199) [Link] My understanding of this is that /dev/urandom reseeds a pseudo random sequence using values from /dev/random as seeds, with a reseeding frequency dependent upon how much genuine entropy is available.
|
|||||||||||||