thx for your answer... i use the 2.6.10 vanilla kernel.. i have downloaded two weeks ago the latest iproute2. I have installed tbf with the following command $tc qdisc add dev $DEV parent 2:1 tbf rate $EF_RATE burst $EF_BURST mtu $EF_MTU limit $EF_LIMIT and i have got the results using tc -s qdisc ls dev eth1 I dont really think that i have done something wrong for taking such calculation errors for the latency... I plan to measure the latency this week with the appropriate scenario.. Tell me if u want to send you a short description of my scenario and of course of my results about the limit and the latency. Thx in advance... I would to contribute for the lartc... that's all --- Andy Furniss <andy.furniss@xxxxxxxxxxxxx> wrote: > Alaios wrote: > > Hi i have found a problem related with tbf and the > > latency that the tbf calculates.. I have used the > > following parameters for burst and limit > > burst 100Kbit limit 500Kbit lat81.8ms > > burst 6Kbit limit 6Kbit lat 0us > > burst 200Kbit limit 100Kbit lat 4294.9s > > What is the full command you used to get those? > > > > > As u can see in the 3rd column the latency for > 100Kbit > > burst and 500Kbit limit is 81.8ms but for 200Kbit > and > > limit 100Kbit is 4294.9s!!! How could be > possible??? > > I want to find a way to caclulate the latency for > a > > packet entering my tbf.. What do u suggest me to > do? > > latency is worked out from rate and limit (you can > also specify latency > and let tbf work out limit (it approximates, > though)). > > > > > Ah! I use the tc -s qdisc ls dev eth1 command for > the > > tbf statistics > > Could be a stats bug that is fixed now, what > kernel/tc version do you use. > > Andy. > > __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/ _______________________________________________ LARTC mailing list LARTC@xxxxxxxxxxxxxxx http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
