3.4. Using the Random Number Generator

In order to be able to generate secure cryptographic keys that cannot be easily broken, a source of random numbers is required. Generally, the more random the numbers are, the better the chance of obtaining unique keys. Entropy for generating random numbers is usually obtained from computing environmental “noise” or using a hardware random number generator.
The rngd daemon, which is a part of the rng-tools package, is capable of using both environmental noise and hardware random number generators for extracting entropy. The daemon checks whether the data supplied by the source of randomness is sufficiently random and then stores it in the kernel's random-number entropy pool. The random numbers it generates are made available through the /dev/random and /dev/urandom character devices.
The difference between /dev/random and /dev/urandom is that the former is a blocking device, which means it stops supplying numbers when it determines that the amount of entropy is insufficient for generating a properly random output. Conversely, /dev/urandom is a non-blocking source, which reuses the kernel's entropy pool and is thus able to provide an unlimited supply of pseudo-random numbers, albeit with less entropy. As such, /dev/urandom should not be used for creating long-term cryptographic keys.
To install the rng-tools package, issue the following command as the root user:
~]# yum install rng-tools
To start the rngd daemon, execute the following command as root:
~]# service rngd start
To query the status of the daemon, use the following command:
~]# service rngd status
To start the rngd daemon with optional parameters, execute it directly. For example, to specify an alternative source of random-number input (other than /dev/hwrandom), use the following command:
~]# rngd --rng-device=/dev/hwrng
The above command starts the rngd daemon with /dev/hwrng as the device from which random numbers are read. Similarly, you can use the -o (or --random-device) option to choose the kernel device for random-number output (other than the default /dev/random). See the rngd(8) manual page for a list of all available options.
The rng-tools package also contains the rngtest utility, which can be used to check the randomness of data. To test the level of randomness of the output of /dev/random, use the rngtest tool as follows:
~]$ cat /dev/random | rngtest -c 1000
rngtest 2
Copyright (c) 2004 by Henrique de Moraes Holschuh
This is free software; see the source for copying conditions.  There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

rngtest: starting FIPS tests...
rngtest: bits received from input: 20000032
rngtest: FIPS 140-2 successes: 1000
rngtest: FIPS 140-2 failures: 0
rngtest: FIPS 140-2(2001-10-10) Monobit: 0
rngtest: FIPS 140-2(2001-10-10) Poker: 0
rngtest: FIPS 140-2(2001-10-10) Runs: 0
rngtest: FIPS 140-2(2001-10-10) Long run: 1
rngtest: FIPS 140-2(2001-10-10) Continuous run: 0
rngtest: input channel speed: (min=308.697; avg=623.670; max=730.823)Kibits/s
rngtest: FIPS tests speed: (min=51.971; avg=137.737; max=167.311)Mibits/s
rngtest: Program run time: 31461595 microseconds
A high number of failures shown in the output of the rngtest tool indicates that the randomness of the tested data is sub-optimal and should not be relied upon. See the rngtest(1) manual page for a list of options available for the rngtest utility.