thr3ads.net - LARTC - R2q stuff [Jun 2004]

If this information is useful, please help other people find it:
Share via:

<listas@dejawu.com.ar>

2004-Jun-11 22:22 UTC

R2q stuff

Hi, im using this script to shape traffic as I want..i get this log on
the kernel 
Log:

HTB: quantum of class 10020 is big. Consider r2q change.
HTB init, kernel part version 3.13
HTB init, kernel part version 3.13

How worryed should I be :) ???

Thanks people!?

#!/bin/bash -x


        fast_mark=1
        ultra_fast_mark=2
        slow_mark=3

        interfaz_externa="eth1"
        interfaz_interna="eth0"
        iptables="/usr/local/sbin/iptables"

        average_port="22 80 8080 3128 443 53 110 5190 259 1041 1723 1863
6667 6668 6669"
        p2p_port="1214 4600:4700 1100:1300 21 6346 3135 3531 1911 21
3995 4242 2780 2527 2152 21"
        p2p_multiport="50000:60000"

        uplink_bw=1524
        downlink_bw=1524

        red_interna1="10.0.0.0/8"
        red_interna3="172.0.3.0/24"
        red_interna2="xxx.xxx.xxx.xxx/27"

        /sbin/iptables -t mangle -F
        /sbin/iptables -t mangle -X

        # ICMP
        $iptables -t mangle -A PREROUTING -p icmp -j MARK --set-mark
$fast_mark

        # TRAFICO OUT
        for i in $average_port
        do
        $iptables -t mangle -N FORW_FAST_OUT_DST_$i
        $iptables -t mangle -A FORWARD -i $interfaz_interna -o
$interfaz_externa -p tcp -m mport --ports $i -j FORW_FAST_OUT_DST_$i
        $iptables -t mangle -A FORW_FAST_OUT_DST_$i -j MARK --set-mark
$fast_mark
        done

        # dns fast
        $iptables -t mangle -A  FORWARD -p udp -m mport --ports 53 -j
MARK --set-mark $fast_mark

        for i in $p2p_port
        do
        $iptables -t mangle -N FORW_P2P_OUT_DST_$i
        $iptables -t mangle -A FORWARD -i $interfaz_interna -o
$interfaz_externa -p tcp -m mport --ports $i -j FORW_P2P_OUT_DST_$i
        $iptables -t mangle -A FORW_P2P_OUT_DST_$i -j MARK --set-mark
$slow_mark
        done


        # TRAFICO IN
        for i in $average_port
        do
        $iptables -t mangle -N FORW_FAST_IN_DST_$i
        $iptables -t mangle -A FORWARD -i $interfaz_externa -o
$interfaz_interna -p tcp -m mport --ports $i -j FORW_FAST_IN_DST_$i
        $iptables -t mangle -A FORW_FAST_IN_DST_$i -j MARK --set-mark
$fast_mark
        done

        for i in $p2p_port
        do
        $iptables -t mangle -N FORW_P2P_IN_DST_$i
        $iptables -t mangle -A FORWARD -o $interfaz_interna -i
$interfaz_externa -p tcp -m mport --ports $i -j FORW_P2P_IN_DST_$i
        $iptables -t mangle -A FORW_P2P_IN_DST_$i -j MARK --set-mark
$slow_mark
        done

        $iptables -t mangle -N FORW_P2P_IN_DST_MULTIPORT
        $iptables -t mangle -A FORWARD -p tcp -m mport --ports
$p2p_multiport -j FORW_P2P_IN_DST_MULTIPORT
        $iptables -t mangle -A FORW_P2P_IN_DST_MULTIPORT -j MARK
--set-mark $slow_mark

        # p2p probar unas coas
        $iptables -t mangle -A PREROUTING -m p2p -j CONNMARK --set-mark
$slow_mark
        $iptables -t mangle -A PREROUTING -m connmark --mark $slow_mark
-j CONNMARK --restore-mark



        # TRAFICO GENERADO POR EL SERVER HACIA INTERNET
        for i in $average_port
        do
        $iptables -t mangle -N OUT_SERVER_OUT_DST_$i
        $iptables -t mangle -A OUTPUT -o $interfaz_externa -p tcp -m
mport --ports $i -j OUT_SERVER_OUT_DST_$i
        $iptables -t mangle -A OUT_SERVER_OUT_DST_$i -j MARK --set-mark
$fast_mark
        done

        # TRAFICO GENERADO POR EL SERVER HACIA LA RED INTERNA
        $iptables -t mangle -A OUTPUT -o $interfaz_interna -j MARK
--set-mark $ultra_fast_mark


        # CLASES PARA EL TRAFICO EXTERNO
        tc qdisc del dev $interfaz_externa root    2> /dev/null >
/dev/null
        tc qdisc del dev $interfaz_externa ingress 2> /dev/null >
/dev/null
        tc qdisc add dev $interfaz_externa root handle 1: htb default 20
        tc class add dev $interfaz_externa parent 1: classid 1:1 htb
rate ${uplink_bw}kbit

        #tc class add dev $interfaz_externa parent 1:1 classid 1:10 htb
rate $[5*$uplink_bw/10]kbit ceil $[6*$uplink_bw/10]kbit prio 1
        #tc class add dev $interfaz_externa parent 1:1 classid 1:20 htb
rate $[3*$uplink_bw/10]kbit ceil $[4*$uplink_bw/10]kbit prio 2
        #tc class add dev $interfaz_externa parent 1:1 classid 1:30 htb
rate $[3*$uplink_bw/10]kbit ceil $[3*$uplink_bw/10]kbit prio 3

        # modificacion viernes 28/2004
        tc class add dev $interfaz_externa parent 1:1 classid 1:10 htb
rate $[5*$uplink_bw/10]kbit ceil $[6*$uplink_bw/10]kbit prio 1
        tc class add dev $interfaz_externa parent 1:1 classid 1:20 htb
rate $[4*$uplink_bw/10]kbit ceil $[4*$uplink_bw/10]kbit prio 2
        tc class add dev $interfaz_externa parent 1:1 classid 1:30 htb
rate $[1*$uplink_bw/10]kbit ceil $[2*$uplink_bw/10]kbit prio 3

        tc qdisc add dev $interfaz_externa parent 1:10 handle 10: sfq
perturb 10
        tc qdisc add dev $interfaz_externa parent 1:20 handle 20: sfq
perturb 10
        tc qdisc add dev $interfaz_externa parent 1:30 handle 30: sfq
perturb 10

        tc filter add dev $interfaz_externa protocol ip prio 1 parent 1:
handle $fast_mark fw classid 1:10
        tc filter add dev $interfaz_externa protocol ip prio 3 parent 1:
handle $slow_mark fw classid 1:30

        tc filter add dev $interfaz_externa parent 1:0 protocol ip prio
1 u32 match ip protocol 1 0xff flowid 1:10


        # CLASES PARA EL TRAFICO INTERNO
        tc qdisc del dev $interfaz_interna root    2> /dev/null >
/dev/null
        tc qdisc del dev $interfaz_interna ingress 2> /dev/null >
/dev/null
        tc qdisc add dev $interfaz_interna root handle 1: htb default 30
        tc class add dev $interfaz_interna parent 1: classid 1:1 htb
rate $[10*$downlink_bw]kbit

        tc class add dev $interfaz_interna parent 1:1 classid 1:10 htb
rate $[5*$downlink_bw/10]kbit ceil $[7*$downlink_bw/10]kbit prio 1

        # I use this because some traffic I don''t want to be shapped
goes on this flow..
       tc class add dev $interfaz_interna parent 1:1 classid 1:20 htb
rate $[80*$downlink_bw]kbit ceil $[80*$downlink_bw]kbit

        tc class add dev $interfaz_interna parent 1:1 classid 1:30 htb
rate $[3*$downlink_bw/10]kbit ceil $[4*$uplink_bw/10]kbit prio 2
        tc class add dev $interfaz_interna parent 1:1 classid 1:40 htb
rate $[1*$downlink_bw/10]kbit ceil $[2*$downlink_bw/10]kbit prio 3

        tc qdisc add dev $interfaz_interna parent 1:10 handle 10: sfq
perturb 10
        tc qdisc add dev $interfaz_interna parent 1:20 handle 20: sfq
perturb 10
        tc qdisc add dev $interfaz_interna parent 1:40 handle 40: sfq
perturb 10

        tc filter add dev $interfaz_interna protocol ip prio 1 parent 1:
handle $fast_mark fw classid 1:10
        tc filter add dev $interfaz_interna protocol ip prio 2 parent 1:
handle $slow_mark fw classid 1:40
        tc filter add dev $interfaz_interna protocol ip prio 2 parent 1:
handle $ultra_fast_mark fw classid 1:20

        tc filter add dev $interfaz_interna parent 1:0 protocol ip prio
1 u32 match ip protocol 1 0xff flowid 1:10

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

Stef Coene

2004-Jun-12 10:17 UTC

head link

Re: R2q stuff

On Saturday 12 June 2004 00:22, listas@dejawu.com.ar
wrote:> Hi, im using this script to shape traffic as I want..i get this log on
> the kernel
> Log:
>
> HTB: quantum of class 10020 is big. Consider r2q change.
> HTB init, kernel part version 3.13
> HTB init, kernel part version 3.13
>
> How worryed should I be :) ???Not so much if everything is working like expected.
Maybe this can help:
http://www.docum.org/docum.org/faq/cache/31.html

Stef

-- 
stef.coene@docum.org
 "Using Linux as bandwidth manager"
     http://www.docum.org/
_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

LARTC - Jun 2004 - R2q stuff

R2q stuff

Re: R2q stuff