Mark Beck
2002-Apr-10 11:52 UTC
[2nd try] Clarification required for Prio in CBQ dequeing order
>Date: Sat, 06 Apr 2002 14:01:33 +0100 >To: lartc@mailman.ds9a.nl >From: Mark Beck <wilson99@lineone.net> >Subject: Clarification required for Prio in CBQ dequeing order > >Hi, > >I have been modifying the Wonder shaper to include another queue. My >understanding was packets would be de queued to hardware in the following >order: > > cbq1 -> sfq10 > sfq20 > sfq30. > >Also we only move onto the next queue when the preceding one is empty. > >However is seems to be prio xx number that reflects the order packets are >de queued not to what flowid they belong to. > >If I enter the commands below and ping my next hop with lots of traffic >matching sport 10240 0x200. > ># ICMP (ip protocol 1) in the interactive class 1:10 so we ># can do measurements & impress our friends: >tc filter add dev $DEV parent 1:0 protocol ip prio 11 u32 \ > match ip protocol 1 0xff flowid 1:10 >tc filter add dev $DEV parent 1:0 protocol ip prio 24 u32 \ > match ip sport 10240 0x200 flowid 1:30 > > Packets: Sent = 85, Received = 85, Lost = 0 (0% loss), >Approximate round trip times in milli-seconds: > Minimum = 20ms, Maximum = 145ms, Average = 79ms > >--------------------------------------------------------------------------------- >!! Now we change the bottom filter to prio4 in flowid 1:30 > ># ICMP (ip protocol 1) in the interactive class 1:10 so we ># can do measurements & impress our friends: >tc filter add dev $DEV parent 1:0 protocol ip prio 11 u32 \ > match ip protocol 1 0xff flowid 1:10 >tc filter add dev $DEV parent 1:0 protocol ip prio 4 u32 \ > match ip sport 10240 0x200 flowid 1:30 > > Packets: Sent = 127, Received = 127, Lost = 0 (0% loss), >Approximate round trip times in milli-seconds: > Minimum = 129ms, Maximum = 2405ms, Average = 475ms > >[root@box root]# tc -s qdisc >qdisc sfq 30: dev ppp0 quantum 1500b perturb 10sec > Sent 6040688 bytes 4199 pkts (dropped 0, overlimits 0) > backlog 34p > > qdisc sfq 20: dev ppp0 quantum 1500b perturb 10sec > Sent 75210 bytes 1167 pkts (dropped 0, overlimits 0) > > qdisc sfq 10: dev ppp0 quantum 1500b perturb 10sec > Sent 317956 bytes 6133 pkts (dropped 0, overlimits 0) > > qdisc cbq 1: dev ppp0 rate 10Mbit (bounded,isolated) prio no-transmit > Sent 6455294 bytes 11521 pkts (dropped 0, overlimits 26113) > backlog 34p > borrowed 0 overactions 0 avgidle 624 undertime 0 > >Packets are sent to the correct flow but ICMP response is far worse! > >--------------------------------------------------------------------------------- > >I know the filters are matching correctly as the right number of packets >go to the correct flow. But could you please confirm if sfq10 should get >de queued before sfq30. > >Thanks, >Mark > > > > >#!/bin/bash > ># The Ultimate Setup For Your Internet Connection At Home ># ># ># Set the following values to somewhat less than your actual download ># and uplink speed. In kilobits >DOWNLINK=800 >UPLINK=220 >DEV=ppp0 > ># clean existing down- and uplink qdiscs, hide errors >tc qdisc del dev $DEV root 2> /dev/null > /dev/null >tc qdisc del dev $DEV ingress 2> /dev/null > /dev/null > >###### uplink > ># install root CBQ > >tc qdisc add dev $DEV root handle 1: cbq avpkt 1000 bandwidth 10mbit > ># shape everything at $UPLINK speed - this prevents huge queues in your ># DSL modem which destroy latency: ># main class > >tc class add dev $DEV parent 1: classid 1:1 cbq rate ${UPLINK}kbit \ >allot 1500 prio 5 bounded isolated > ># high prio class 1:10: > >tc class add dev $DEV parent 1:1 classid 1:10 cbq rate ${UPLINK}kbit \ > allot 1600 prio 1 avpkt 1000 > ># bulk and default class 1:20 - gets slightly less traffic, ># and a lower priority: > >tc class add dev $DEV parent 1:1 classid 1:20 cbq rate $[9*$UPLINK/10]kbit \ > allot 1600 prio 2 avpkt 1000 > ># ftp in 1:30: We send this shit last :) > >tc class add dev $DEV parent 1:1 classid 1:30 cbq rate $[9*$UPLINK/10]kbit \ > allot 1600 prio 3 avpkt 1000 > ># all get Stochastic Fairness: >tc qdisc add dev $DEV parent 1:10 handle 10: sfq perturb 10 >tc qdisc add dev $DEV parent 1:20 handle 20: sfq perturb 10 >tc qdisc add dev $DEV parent 1:30 handle 30: sfq perturb 10 > ># To speed up downloads while an upload is going on, put ACK packets in ># the interactive class: > >tc filter add dev $DEV parent 1: protocol ip prio 10 u32 \ > match ip protocol 6 0xff \ > match u8 0x05 0x0f at 0 \ > match u16 0x0000 0xffc0 at 2 \ > match u8 0x10 0xff at 33 \ > flowid 1:10 > > > ># start filters > ># TOS Minimum Delay (ssh, NOT scp) in 1:10: >#tc filter add dev $DEV parent 1:0 protocol ip prio 10 u32 \ ># match ip tos 0x10 0xff flowid 1:10 > ># ICMP (ip protocol 1) in the interactive class 1:10 so we ># can do measurements & impress our friends: >tc filter add dev $DEV parent 1:0 protocol ip prio 11 u32 \ > match ip protocol 1 0xff flowid 1:10 > ># Halflife >tc filter add dev $DEV parent 1:0 protocol ip prio 12 u32 \ > match tcp src 27005 0xffff flowid 1:10 > >tc filter add dev $DEV parent 1:0 protocol ip prio 12 u32 \ > match udp src 27005 0xffff flowid 1:10 > ># tcp 22 (ssh) >tc filter add dev $DEV parent 1:0 protocol ip prio 13 u32 \ > match tcp src 22 0xffff flowid 1:10 > > ># tcp 23 (telnet) >tc filter add dev $DEV parent 1:0 protocol ip prio 13 u32 \ > match tcp src 23 0xffff flowid 1:10 > > ># UDP >tc filter add dev $DEV parent 1:0 protocol ip prio 14 u32 \ > match ip protocol 17 0xff flowid 1:10 > ># ftp filter IP src ports 10240-10751 > >tc filter add dev $DEV parent 1:0 protocol ip prio 24 u32 \ > match ip sport 10240 0x200 flowid 1:30 > ># rest is ''non-interactive'' ie ''bulk'' and ends up in 1:20. > >tc filter add dev $DEV parent 1: protocol ip prio 25 u32 \ > match ip dst 0.0.0.0/0 flowid 1:20 > >[root@box root]#