On Mon 26 Mar 2007, Henrik Nordstrom wrote: Hi Henrik, > So what was the hardest part to understand? (just trying to figure out > how to better document this) > > http://wiki.squid-cache.org/SquidFaq/MiscFeatures#head-fd9b4b7ba1854a3c2179 >6173af9d0b9aee33e376 > http://www.deckle.co.za/squid-users-guide/Access_Control_and_Access_Control >_Operators#Delay_Classes Basically the terminology (pools, buckets, classes), but I suspect that if I had read it in more quiet environment (at home for example) I wouldn't have had any problems... Actually, now that I look at it again it makes more sense than last week :) > Looks fine if you want a hard 100KByte/s (ca 1Mbit/s) limit per user. > Usually how most people use delay pools is by using a somewhat bigger > bucket (maybe 10 times the rate) and a slightly smaller rate to allow > some burstiness which is typical in normal browsing while still limiting > downloads. > > delay_parameters 1 -1/-1 400000/90000 > > this will allow users to access Internet in bursts of up to 400Kbyte > unlimited as long as they stay within their allowed 90Kbyte/s limit in > average. > > You could also reorder the acl lists slightly to make them easier to > maintain if you need to add more exclusion criterias later on > > delay_access 1 deny fastusers > delay_access 1 deny local > delay_access 1 allow all Right, I say. That's perhaps a better solution... I'll try this instead to see what happens. Thanks, Ray -- You cannot discover new oceans unless you have the courage to lose sight of the shore
