search for: drr

Hi, I used HTB3.6 on Linux to control traffic.It seems that HTB uses DRR(Dificit Round-Robin) algorithm.I heard that it is difficult for DRR to get a delay bounded and WFQ is better than DRR.Is that right?So,how to implement WFQ in Linux? Thanks a lot! Best Regards swcims swcims@163.com 2004-08-03 _______________________________________...

...- are they needed at all? - They do not apply cean. The kernel is 2.6.13. This configuration does not work: CONFIG_IP_ROUTE_MULTIPATH=y CONFIG_IP_ROUTE_MULTIPATH_CACHED=y CONFIG_IP_ROUTE_MULTIPATH_RR=m CONFIG_IP_ROUTE_MULTIPATH_RANDOM=m CONFIG_IP_ROUTE_MULTIPATH_WRANDOM=m CONFIG_IP_ROUTE_MULTIPATH_DRR=m How should i use these modules? ip route add default mpath (rr| drr|random|wrandom)... does not load the appropriate module automatically. Loading them manualy changes nothing. And issuing this command on kernel where this feature is not enebled does not report any error. Krzysiek ----------...

Hi there! I''ve using HTB for a while and now I an faced with a ''problem''. How hierarchical is HTB? Let''s say I have this 3 layer HTB setup: root class 1: (rate=100, ceil=100) 1: children classes 1:10 (30,100) and 1:20 (70,100) 1:10 children classes 1:100 (10,100) and 1:101 (20,100) 1:20 children classes 1:200 (30,100) and 1:201 (70,100) I managed to have

Hello again, I need to change the R2Q for my script, as setting the quantum manually for each class is painful. Can you tell me exactly where to set R2Q = x? I get syntax errors all the time. Should it be specified for each class? I do not know where to place this setting... Thanks in advance, Vlad Mihai _______________________________________________ LARTC mailing list /

...lartc-owner@mailman.ds9a.nl When replying, please edit your Subject line so it is more specific than "Re: Contents of LARTC digest..." Today''s Topics: 1. Re: Traffic Control (Andreas Klauer) 2. Re: LARTC Digest, Vol 7, Issue 6 (Gobbledegeek) 3. ip route add default mpath (rr| drr|random|wrandom)... (Krzysiek) ---------------------------------------------------------------------- Message: 1 Date: Sun, 25 Sep 2005 12:09:52 +0200 From: Andreas Klauer Subject: Re: [LARTC] Traffic Control To: lartc@mailman.ds9a.nl Message-ID: <200509251209.52461.Andreas.Klauer@metamorpher...

Hello, I have been searching for HTB theory documentation and found two interesting sources - Devik''s page and docum.org FAQ. In some places they are contradictory which make me think that Devik''s theoretic document (marked "actual") is just too old and does not reflect the latest changes. After reading http://www.docum.org/docum.org/faq/cache/30.html and others, I

...lartc-owner@mailman.ds9a.nl When replying, please edit your Subject line so it is more specific than "Re: Contents of LARTC digest..." Today''s Topics: 1. Re: Traffic Control (Andreas Klauer) 2. Re: LARTC Digest, Vol 7, Issue 6 (Gobbledegeek) 3. ip route add default mpath (rr| drr|random|wrandom)... (Krzysiek) ---------------------------------------------------------------------- Message: 1 Date: Sun, 25 Sep 2005 12:09:52 +0200 From: Andreas Klauer Subject: Re: [LARTC] Traffic Control To: lartc@mailman.ds9a.nl Message-ID: <200509251209.52461.Andreas.Klauer@metamorpher...

I have this problem: I have an Internet line input with variable speed. I have a max speed and a min speed: Vmax and Vmin. The speed is always changing between Vmax and Vmin. I want to share the actual bandwidth (you don''t not how much, you only know the speed is between Vmax and Vmin) for N clients. The bandwidth should be shared so nobody can get more bandwidth than the others.

Does anyone use tb.init TIMECHECK to change rate/ceil for classes based on periods over the day ? I had problems setting the right syntax. Can anyone help me with a quick example. I have tried the following with no success: LEAF=sfq RULE=xx.xx.xx.xx TIME=06:00-18:00;128Kbit TIME=18:00-06:00;256Kbit RATE=196Kbit _______________________________________________ LARTC mailing list

....0.1 dev eth0 default via 10.0.0.1 dev eth1 metric 1 I enabled ip_forward and set arp_ignore to 1 for eth0 and eth1 to make sure the correct nic answers to arp requests. I tried to get the egress load balancing to work by replacing the above two default routes with: ip route add default mpath drr nexthop via 10.0.0.1 dev eth0 weight 1 onlink nexthop via 10.0.0.1 dev eth1 weight 1 onlink I assumed that with mpath device round robin both nics would be used more or less equally, but the reality is only one of the nics actually works and the second nic even stops responding to arp requests....

...omething that no howto I saw mentioned so far: multipath algorithms. The kernel has the followings: # zgrep MULTIPATH_ /proc/config.gz CONFIG_IP_ROUTE_MULTIPATH_CACHED=y CONFIG_IP_ROUTE_MULTIPATH_RR=m CONFIG_IP_ROUTE_MULTIPATH_RANDOM=m CONFIG_IP_ROUTE_MULTIPATH_WRANDOM=m CONFIG_IP_ROUTE_MULTIPATH_DRR=m CONFIG_DM_MULTIPATH_EMC iproute2 also has support for these (at least, it passed them forward to the kernel): static char *mp_alg_names[IP_MP_ALG_MAX+1] = { [IP_MP_ALG_NONE] = "none", [IP_MP_ALG_RR] = "rr", [IP_MP_ALG_DRR] = "drr", [...

I would like to implement an architecture to support LLQ and WFQ with Qos linux elements ! Is possible use CBQ (you can set same parameters to control WRR process) or HTB (how can i controll DRR process) structure ? My idea is to classify packets by List flows, tos flows or connections flows (as CBWFQ on CISCO ) I saw also for WRR patch (http://www.zz9.dk/wrr), but I don''t understand well how configure parameters to set bandwidth . I saw also for WFQ patch (http://home.sch.bme.h...

...e(Dnlink)?(e.g. why put the police rule on the smaller bandwidth side but not put on the larger side) 2. I can''t differ the functionalities exactly between SFQ and HTB , they''re both the queueing disciplines,HTB is used to manage bandwidth ,with the DRR displine in it , also be a scheduler to send packets, so, why we still need SFQ? 3. Does anybody know that where I can find the VoIP and VOD traffic generator ? ''cause I want to evaluate the performance in the linux DiffServ I don''t know if the iperf can ge...

Dear all, I''m reading codes in sch_sqf.c in kernel 2.4 of Linux. I find I messed up with the concept of slot, bucket and hash tables. Could anyone explain the data stucture of the hash tables a little to me. Thanks very much. Best, Franklin BTW: I can not understand why the number of entries of dep[SFQ_DEPTH*] should be twice:( _______________________________________________ LARTC

...2.6 kernel from the start to finish (ethernet interface up to htb and gred classification). In the sch_htc.c i have found the following /* debugging support; S is subsystem, these are defined: 0 - netlink messages 1 - enqueue 2 - drop & requeue 3 - dequeue main 4 - dequeue one prio DRR part 5 - dequeue class accounting 6 - class overlimit status computation 7 - hint tree 8 - event queue 10 - rate estimator 11 - classifier 12 - fast dequeue cache L is level; 0 = none, 1 = basic info, 2 = detailed, 3 = full q->debug uint32 contains 16 2-bit fields one for subsystem...

Hi All, I have a data frame: Time1 Time2 1 176:46:10 41:48:06 2 171:28:57 61:19:10 3 178:25:15 34:05:35 4 74:04:20 25:01:55 5 136:11:20 37:59:32 6 138:17:17 30:22:27 7 183:04:48 29:25:02 8 179:35:01 19:29:44 > str(df) 'data.frame': 8 obs. of 2 variables: $ Time1: Factor w/ 583 levels

...TCA_HTB_RTAB, + TCA_HTB_NOHYST, __TCA_HTB_MAX, }; diff -Nurp iproute2.orig/tc/q_htb.c iproute2/tc/q_htb.c --- iproute2.orig/tc/q_htb.c 2006-06-13 11:53:27.000000000 +1000 +++ iproute2/tc/q_htb.c 2006-06-13 11:54:50.000000000 +1000 @@ -35,7 +35,7 @@ static void explain(void) " r2q DRR quantums are computed as rate in Bps/r2q {10}\n" " debug string of 16 numbers each 0-3 {0}\n\n" "... class add ... htb rate R1 [burst B1] [mpu B] [overhead O] [atm]\n" - " [prio P] [slot S] [pslot PS]\n" + "...

> body#rails_outertrack_com { > font-size: 14px /* Be kind to these li''l eyes of yours, baby */ > } > > Dominique Rose-Rosette Be kinder by specifying {font-size 1em} so that it''s scalable. My pixels are smaller than yours baby. Seriously though folks, please try and move away from absolute sizes for elements (px). Unless they''ve fixed it IE

Ok, this has been kicking my @$$ for weeks. I'm trying to get some kind of bandwidth shaping working on my server. I need to throttle ftp down so as not to suck up all the available bandwidth. I had cbq working on the old server (an ancient RH 6.2 box) so I figured I'd just move the config over and get cbq.init from sf.net and it should work. Unfortunately it doesn't. At least

Hi again, I just ran into a seriously strange glitch, and was wondering if anybody had experienced this... I have an HTB class with an iptables classifier. The iptables classifier shows packets hitting the rule for this class. Now, HTB drops these packets (''dropped'' increases), even although the packets are small, and there are over 10000 tokens available. Is there any