On Tuesday 22 July 2003 03:12, tanxuey wrote: > Hi all!!! > > > I do not know why my letter was sent back again and again!!! Maybe because there are a lot of ! in the subject and not everybody is appreciating that. > Trever said my ceil value was wrong!!! If you have a 10 mbit and your ceil is 10mbit, you will not shape any traffic because the network is the bottlenect. > My test enviroment is local ethernet network, computer 18 has 10Mbit > network card, and HTB+BRIDGE box has two 100Mbit network card, > > computer 29 and 26 have 100Mbit network card too. 26 and 29 link to > HTB+BRIDGE eth0 by 100Mbit SWITCHER not HUB. > > My ceil parameter set to 10Mbit wrong? > > Yestoday, I found the performace of HTB is related to r2q value and ceil > value. If I set ceil to 100kbps,set r2q to 1,it is ok. > If I set ceil to 200kbps,set r2q to 1.tc always say "HTB: quantum of class > 10001 is big, Consider r2q change". > > In my test environment that I first mentioned,I saw the same information. > > Can you tell me their relation?? See faq page on www.docum.org Stef -- stef.coene@xxxxxxxxx "Using Linux as bandwidth manager" http://www.docum.org/ #lartc @ irc.oftc.net