Hello, Gustavo, thank you for your answer. Am Montag, 31. Juli 2006 00:52 schrieb Gustavo Homem: > On Sun, Jul 30, 2006 at 11:09:04PM +0200, Daniel Musketa wrote: [...] > > This qdisc only affects outgoing traffic. But I also want to control > > incoming packets and keep the isp's queue empty. > > > > Question 2: What is the best solution for doing this: ingress qdisc, IMQ, > > ... (thers's only one link to isp)? > > If you don't want to patch the kernel and your machine has only two network > interfaces you can shape the outgoing traffic to the internal interface > instead of the incoming traffic to the internal interface. I have an > example script here: > > http://downloads.angulosolido.pt/QoS/HTB_shaper_adv.sh > > It uses HTB for shapping though. Sorry, I should have explained a little more detailed: (Nearly) all download traffic is initiated by the machine itself, so I can't use the "limiting the clients" trick here. > > Much traffic on this line is UDP traffic (OpenVPN). > > > > Question 3: If I do policing on incoming traffic, do UDP transmissions > > care about dropping and reduce transmit speed? > > I don't think so, unless that is implemented at the application level. The > UDP packets shouldn't be dropped unless that's acceptable to a certain > extend (VoIP) or the application resends as necessary. I am sure OpenVPN resends dropped packets (sooner or later) but don't know yet if it adapts its transmit rate to the rate of the reply packets ... > > If I begin to control incoming traffic I only want to drop packets that > > are non voip traffic. > > > > Question 4: Is it possible to control what packets are dropped? > > Depends if you know the ports they use. If you know it, you just have to > mark them accordingly. > > For example with skype it's a pain in the ass, The only voip traffic (over ppp0) is produced by an asterisk server (on the same machine). I know all used ports (SIP and RTP) and for outgoing packets I already mark them with iptables. > > My idea: A script that periodically (or even better: triggered by > > asterisk) controls the parameters for the ingress shaper (depending on > > the actual upload traffic produced by voip which could easily be measured > > with tc). > > > > Question 5: Is that possible? Is this necessary at all? Is there already > > a solution? > > I don't think you need this. A setup with HTB solves this problem, since > each traffic class has a defined RATE as well as a defined CEIL rate, which > it will take whenever available. But then I need IMQ and a patched kernel, right? > Best regards > Gustavo Muito obrigado pela sua ajuda ... Daniel _______________________________________________ LARTC mailing list LARTC@xxxxxxxxxxxxxxx http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc
