similar to: Re: Is this actually possible?

Hi, I''m trying to setup QoS in my TSL box. Following to mailing list reccomendations I changed PSCHED_CPU but can''t get accurate shaping with HTB. I have disabled HYSTERESIS also (by setting it to 0). I have Intel Celeron 1.8Mhz (TSC supported I think). ******** My script ****** tc qdisc add dev eth0 root handle 1: htb default 22 r2q 10 # Class corresponding interface

hi i ve seen one of your replies here and i saw that you receomend 2 modifications in the kernel files in order to improve htb. first in pkt_sched.h, changing PSCHED_JIFFIES to PSCHED_CPU. and then, sch_htb.c, changing HYSTERESIS from 1 to 0. is it that simple? i mean i just change these here variables and i get better results? i ve done a search and found the first file in 3 locations. where

Hello good day to all ... this is my setup 1 Linux Wireless Access Point, connected are 4 wireless gateway in which i needed to apply shaping ... ok here is the weird part... clients on each gateway download files from the Acess Point ... a 500 mb file through ftp on gateway 1 which is up to 64 kbps ... the result is from 60-64 kbps speed which is fine ... on gateway 2 which is 128 kbps ... the

Hello LARTC, Internet connection - ADSL, interface nas0 (115kbit guarantied, up to 1mbit possible. Depends on ISP load, impossible to guess). Internal interface LAN, eth0. Is it possible to successfully shape download traffic on eth0 using HTB? Classes must have guarantied rate calculated from 115kbit possible rate (for example 3 classes) and the possibility to borrow up to 1mbit (depends on ISP

Hey, fellow LART controlllers, I''d like to ask if someome has a real esfq experience with HTB class hierarchy and esfq qdiscs. Lets say that I want to unite the connections to 10 client PCs with u32 or fw filters (doesn''t matter which one in fact) into one HTB class with esfq qdisc at the end. Did anyone tried it, and more important, did it worked the way it is supposed to? As

Hello everyone: As it is known, when you limit uplink bandwidth it usually gets downlink bandwidth to a lower value. I just want to know what is the optimal configuration for eth1 and imq0 according to some variables of tc(HTB), txqueuelen, mtu, etc. to make these packet flows less independent on an ethernet based network. Thank you in advance. -Diego

Hi! I have a setup where I want to prefer traffic on one port (for testing purposes I used port 22) my setup is : tc qdisc add dev eth3 root handle 1: htb default 30 tc class add dev eth3 parent 1: classid 1:1 htb rate 96mbit burst 15k tc class add dev eth3 parent 1: classid 1:7 htb rate 2mbit burst 15k tc class add dev eth3 parent 1:1 classid 1:10 htb rate 96mbit burst 15k tc class add dev

Hello one more time, As others seem to already asked without reply, I''m getting lower speed rates than specified via ingress. How do I know. Because I have this in my script: tc qdisc add dev $DEV handle ffff: ingress # Filter intranet traffic, so fit it to intranet speed tc filter add dev $DEV parent ffff: protocol ip prio 10 u32 \ match ip src $Q_2_HOSTS \ match ip dst

Zviad O. Giorgadze wrote: > My ISP provides guarantied 115kbit bandwidth for GLOBAL TRAFFIC. During the low load period (early morning, evening, night) customers can get up to 1mbit traffic. That''s download traffic we''re talking about, since you seem to be shaping on your local LAN interface? Variable rate ISPs are tough to shape right, I guess... Does this 115kbit vs.

HTB: class htb 1:40 parent 1:1 leaf 40: prio 3 rate 358Kbit ceil 529Kbit \ burst 6Kb cburst 2260b Sent 145871726 bytes 97293 pkts (dropped 69, overlimits 0) rate 56741bit 37pps backlog 23p lended: 77429 borrowed: 19841 giants: 0 I would like to increase "backlog" because I think that would decrease "dropped". 23 packets of 1500 bytes each is only 34,500 bytes. IMO, there

Is there an other place than www.docum.org where one could get monitor_tc.pl? Docum.org seems to be down. Peter -- The bogosity meter just pegged.

Hi again, sorry to be such a bother. I got my setup to work with kernel 2.6.8.1, however the two machines where I need to implement the shaping are running a 2.4.27 kernel. I have applied the infamous June patch (htbfair.diff) already, and recompiled the modules. And I am using the tc that comes with htb3.6-020525.tgz. While I can see the packets going into the right class, it does not

Because these tend to get lost, here are some links that I think everyone should know about: wiki: http://linux-net.osdl.org API: http://www.coverfire.com/lql/ Stef Coene: (Broken? I hope not. This is excellent stuff!) http://www.docum.org/docum.org/ Jason Boxman: http://edseek.com/~jasonb/articles/traffic_shaping/ Dan Singletary:

I posted earlier when I noticed that htb was releasing packets in pairs, even though my burst/quantums were 1 pkt. To fix I set HTB_HYSTERESIS 0 in net/sched/sch_htb.c . This gives a noticable gain in upstream worst case latency, for me with 256kbit/s up I used to see +90 sometimes, now it''s +45. For the many who have 128 up it should limit them to +90 rather than +180. Andy.

does anyone know a way to shape outgoing/upload traffic per ip? I have a network and i want to limit the upload with 100kbit per user. Ex: 192.168.1.20 ----> 1024kbit-DOWN / 100kbit-UP 192.168.1.21 ----> 1024kbit-DOWN / 100kbit-UP and so on....... Ive tried CBQ and HTB, but couldnt get is right. the only thing that I did in upload bases was: "tc qdisc add dev ppp0 root tbf rate

Hello, I have a serious problem with HTB which I wasn''t able to solve myself. I run a masquerading router with ppp0 as interface to the Internet. Three clients need to share a downstream of 1 MBit, which I want to divide with tc. When I see a packet being forwarded to one of these clients, I give it the appropriate unique mark: iptables -t mangle -A FORWARD -d 192.168.34.141 -j MARK

The docs[1][2] suggest it''s 1024, but tc says something else: # tc qdisc add dev eth0 root tbf rate 1kbps latency 50ms burst 1500 # tc -s qdisc ls dev eth0 qdisc tbf 8009: rate 8000bit burst 1499b lat 48.8ms ^^^^^^^ Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) rate 0bit 0pps backlog 0b 0p requeues 0 If 1k were 1024, then I would have 8192bit above.

The manual says (about prio): > Because it doesn''t actually shape, the same warning as for SFQ holds: > either use it only if your physical link is really full or wrap it > inside a classful qdisc that does shape. The latter holds for almost all > cable modems and DSL devices. I want to wrap prio inside of tbf. Here''s why: I have a server on a DSL line, which has

In this simple htb setup: # tc -s -d class ls dev eth0 class htb 1:1 root rate 300000bit ceil 300000bit burst 1749b/8 mpu 0b overhead 0b cburst 1749b/8 mpu 0b overhead 0b level 7 Sent 13171835 bytes 13169 pkt (dropped 0, overlimits 0 requeues 0) rate 45848bit 10pps backlog 0b 0p requeues 0 lended: 5272 borrowed: 0 giants: 0 tokens: -84429 ctokens: -84429 class htb 1:2 parent 1:1 prio 0

Hi, New to traffic shaping I would like to make some tests. It''s for a production server (fw), using 2.4.27 kernel. Something is not clear to me about HTB... is the last stable version included in the 2.4.27 kernel and iproute2 (iproute2-2.6.8-040823.tar.bz2) ? Are there special patches to apply ? EC. _______________________________________________ LARTC mailing list /