doc/actions/ifb-README

   1
   2 IFB is intended to replace IMQ.
   3 Advantage over current IMQ; cleaner in particular in in SMP;
   4 with a _lot_ less code.
   5
   6 Known IMQ/IFB USES
   7 ------------------
   8
   9 As far as i know the reasons listed below is why people use IMQ.
  10 It would be nice to know of anything else that i missed.
  11
  12 1) qdiscs/policies that are per device as opposed to system wide.
  13 IFB allows for sharing.
  14
  15 2) Allows for queueing incoming traffic for shaping instead of
  16 dropping. I am not aware of any study that shows policing is
  17 worse than shaping in achieving the end goal of rate control.
  18 I would be interested if anyone is experimenting.
  19
  20 3) Very interesting use: if you are serving p2p you may want to give
  21 preference to your own locally originated traffic (when responses come back)
  22 vs someone using your system to do bittorent. So QoSing based on state
  23 comes in as the solution. What people did to achieve this was stick
  24 the IMQ somewhere prelocal hook.
  25 I think this is a pretty neat feature to have in Linux in general.
  26 (i.e not just for IMQ).
  27 But i won't go back to putting netfilter hooks in the device to satisfy
  28 this.  I also don't think its worth it hacking ifb some more to be
  29 aware of say L3 info and play ip rule tricks to achieve this.
  30 --> Instead the plan is to have a conntrack related action. This action will
  31 selectively either query/create conntrack state on incoming packets.
  32 Packets could then be redirected to ifb based on what happens -> eg
  33 on incoming packets; if we find they are of known state we could send to
  34 a different queue than one which didn't have existing state. This
  35 all however is dependent on whatever rules the admin enters.
  36
  37 At the moment this 3rd function does not exist yet. I have decided that
  38 instead of sitting on the patch for another year, to release it and then
  39 if there is pressure i will add this feature.
  40
  41 An example, to provide functionality that most people use IMQ for below:
  42
  43 --------
  44 export TC="/sbin/tc"
  45
  46 $TC qdisc add dev ifb0 root handle 1: prio
  47 $TC qdisc add dev ifb0 parent 1:1 handle 10: sfq
  48 $TC qdisc add dev ifb0 parent 1:2 handle 20: tbf rate 20kbit buffer 1600 limit 3000
  49 $TC qdisc add dev ifb0 parent 1:3 handle 30: sfq
  50 $TC filter add dev ifb0 protocol ip pref 1 parent 1: handle 1 fw classid 1:1
  51 $TC filter add dev ifb0 protocol ip pref 2 parent 1: handle 2 fw classid 1:2
  52
  53 ifconfig ifb0 up
  54
  55 $TC qdisc add dev eth0 ingress
  56
  57 # redirect all IP packets arriving in eth0 to ifb0
  58 # use mark 1 --> puts them onto class 1:1
  59 $TC filter add dev eth0 parent ffff: protocol ip prio 10 u32 \
  60 match u32 0 0 flowid 1:1 \
  61 action ipt -j MARK --set-mark 1 \
  62 action mirred egress redirect dev ifb0
  63
  64 --------
  65
  66
  67 Run A Little test:
  68
  69 from another machine ping so that you have packets going into the box:
  70 -----
  71 [root@jzny action-tests]# ping 10.22
  72 PING 10.22 (10.0.0.22): 56 data bytes
  73 64 bytes from 10.0.0.22: icmp_seq=0 ttl=64 time=2.8 ms
  74 64 bytes from 10.0.0.22: icmp_seq=1 ttl=64 time=0.6 ms
  75 64 bytes from 10.0.0.22: icmp_seq=2 ttl=64 time=0.6 ms
  76
  77 --- 10.22 ping statistics ---
  78 3 packets transmitted, 3 packets received, 0% packet loss
  79 round-trip min/avg/max = 0.6/1.3/2.8 ms
  80 [root@jzny action-tests]#
  81 -----
  82 Now look at some stats:
  83
  84 ---
  85 [root@jmandrake]:~# $TC -s filter show parent ffff: dev eth0
  86 filter protocol ip pref 10 u32
  87 filter protocol ip pref 10 u32 fh 800: ht divisor 1
  88 filter protocol ip pref 10 u32 fh 800::800 order 2048 key ht 800 bkt 0 flowid 1:1
  89   match 00000000/00000000 at 0
  90         action order 1: tablename: mangle  hook: NF_IP_PRE_ROUTING
  91         target MARK set 0x1
  92         index 1 ref 1 bind 1 installed 4195sec  used 27sec
  93          Sent 252 bytes 3 pkts (dropped 0, overlimits 0)
  94
  95         action order 2: mirred (Egress Redirect to device ifb0) stolen
  96         index 1 ref 1 bind 1 installed 165 sec used 27 sec
  97          Sent 252 bytes 3 pkts (dropped 0, overlimits 0)
  98
  99 [root@jmandrake]:~# $TC -s qdisc
 100 qdisc sfq 30: dev ifb0 limit 128p quantum 1514b
 101  Sent 0 bytes 0 pkts (dropped 0, overlimits 0)
 102 qdisc tbf 20: dev ifb0 rate 20Kbit burst 1575b lat 2147.5s
 103  Sent 210 bytes 3 pkts (dropped 0, overlimits 0)
 104 qdisc sfq 10: dev ifb0 limit 128p quantum 1514b
 105  Sent 294 bytes 3 pkts (dropped 0, overlimits 0)
 106 qdisc prio 1: dev ifb0 bands 3 priomap  1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1
 107  Sent 504 bytes 6 pkts (dropped 0, overlimits 0)
 108 qdisc ingress ffff: dev eth0 ----------------
 109  Sent 308 bytes 5 pkts (dropped 0, overlimits 0)
 110
 111 [root@jmandrake]:~# ifconfig ifb0
 112 ifb0    Link encap:Ethernet  HWaddr 00:00:00:00:00:00
 113           inet6 addr: fe80::200:ff:fe00:0/64 Scope:Link
 114           UP BROADCAST RUNNING NOARP  MTU:1500  Metric:1
 115           RX packets:6 errors:0 dropped:3 overruns:0 frame:0
 116           TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
 117           collisions:0 txqueuelen:32
 118           RX bytes:504 (504.0 b)  TX bytes:252 (252.0 b)
 119 -----
 120
 121 You send it any packet not originating from the actions it will drop them.
 122 [In this case the three dropped packets were ipv6 ndisc].
 123
 124 cheers,
 125 jamal