On Wed, 6 Feb 2008, Stephen Gran wrote:
> On Wed, Feb 06, 2008 at 03:24:17PM +0000, W B Hacker said:
> >
> > - absent a stream of test traffic, are existing tools in fact adequate
> > to 'predict' what a change to ratelimit settings 'should be expected' to
> > produce as a result?
>
> I am not sure how you hope to be able to predict a stochastic process
> without at least attempting to model it. Am I missing something?
There is a small amount of information in the spec about how the rate
measurement works. I deliberately did not go into detail because I did not
want the maths to be off-putting, and in any case I chose the model
carefully so that it would behave in a fairly intuitive manner. If you
want a better idea of how it works, look at
http://www-uxsup.csx.cam.ac.uk/~fanf2/hermes/doc/antiforgery/ratelimit.html
There's also a brief summary in the code.
If you want to experiment with the model, try piping data into the
util/ratelimit.pl script.
Tony.
--
<fanf@???> <dot@???>
http://dotat.at/ ${sg{\N${sg{\
N\}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}\
\N}{([^N]*)(.)(.)(.*)}{\$1\$3\$2\$1\$3\n\$2\$3\$4\$3\n\$3\$2\$4}}
