I forgot to mention that I've also started using the suggestion someone
made earlier of increasing failureTableSize. I'm now using the formula:
failureTableSize = maxRequestsPerInterval *
(failureTableTime / maxRequestsInterval)
(too bad the config parser can't handle this sort of thing) :-)
Seems like a good idea, as within any failureTableTime period, I'm often
seeing more than the default failureTableSize number of failures.
I've started using a similar approach in a number of places, e.g.:
lsMaxTableSize=512 (maxNodeConnections)
lsHalfLifeHours=0.1
lsAcceptRatioSamples=1800 (maxRequestsPerInterval * 60 * lsHalfLifeHours)
I think making certain settings dependent on other settings is a wise
policy, and makes the configuration just feel more "consistent".
Seems to work better that way, too. :-)
--
Conrad Sabatier <[EMAIL PROTECTED]> - "In Unix veritas"
_______________________________________________
Devl mailing list
[EMAIL PROTECTED]
http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
