tc qdisc ... dev dev ( parent classid | root) [ handle major: ] cbq [
allot bytes ] avpkt bytes bandwidth rate [ cell bytes ] [ ewma log ] [
mpu bytes ]
tc class ... dev dev parent major:[minor] [ classid major:minor ] cbq
allot bytes [ bandwidth rate ] [ rate rate ] prio priority [ weight
weight ] [ minburst packets ] [ maxburst packets ] [ ewma log ] [ cell
bytes ] avpkt bytes [ mpu bytes ] [ bounded isolated ] [ split handle &
defmap defmap ] [ estimator interval timeconstant ]
Class Based Queueing is a classful qdisc that implements a rich
linksharing hierarchy of classes. It contains shaping elements as well
as prioritizing capabilities. Shaping is performed using link idle
time calculations based on the timing of dequeue events and underlying
When shaping a 10mbit/s connection to 1mbit/s, the link will be idle
90% of the time. If it isn't, it needs to be throttled so that it IS
idle 90% of the time.
During operations, the effective idletime is measured using an exponen-
tial weighted moving average (EWMA), which considers recent packets to
be exponentially more important than past ones. The Unix loadaverage is
calculated in the same way.
The calculated idle time is subtracted from the EWMA measured one, the
resulting number is called 'avgidle'. A perfectly loaded link has an
avgidle of zero: packets arrive exactly at the calculated interval.
An overloaded link has a negative avgidle and if it gets too negative,
CBQ throttles and is then 'overlimit'.
Conversely, an idle link might amass a huge avgidle, which would then
allow infinite bandwidths after a few hours of silence. To prevent
this, avgidle is capped at maxidle.
If overlimit, in theory, the CBQ could throttle itself for exactly the
amount of time that was calculated to pass between packets, and then
pass one packet, and throttle again. Due to timer resolution con-
straints, this may not be feasible, see the minburst parameter below.
Within the one CBQ instance many classes may exist. Each of these
classes contains another qdisc, by default tc-pfifo(8).
When enqueueing a packet, CBQ starts at the root and uses various meth-
ods to determine which class should receive the data.
are done. Otherwise, restart.
(ii) Consult the defmap for the priority assigned to this packet,
which depends on the TOS bits. Check if the referral is leaf-
less, otherwise restart.
(iii) Ask the defmap for instructions for the 'best effort' priority.
Check the answer for leafness, otherwise restart.
(iv) If none of the above returned with an instruction, enqueue at
This algorithm makes sure that a packet always ends up somewhere, even
while you are busy building your configuration.
For more details, see tc-cbq-details(8).
LINK SHARING ALGORITHM
When dequeuing for sending to the network device, CBQ decides which of
its classes will be allowed to send. It does so with a Weighted Round
Robin process in which each class with packets gets a chance to send in
turn. The WRR process starts by asking the highest priority classes
(lowest numerically - highest semantically) for packets, and will con-
tinue to do so until they have no more data to offer, in which case the
process repeats for lower priorities.
Classes by default borrow bandwidth from their siblings. A class can be
prevented from doing so by declaring it 'bounded'. A class can also
indicate its unwillingness to lend out bandwidth by being 'isolated'.
The root of a CBQ qdisc class tree has the following parameters:
parent major:minor | root
This mandatory parameter determines the place of the CBQ
instance, either at the root of an interface or within an exist-
Like all other qdiscs, the CBQ can be assigned a handle. Should
consist only of a major number, followed by a colon. Optional,
but very useful if classes will be generated within this qdisc.
This allotment is the 'chunkiness' of link sharing and is used
for determining packet transmission time tables. The qdisc allot
differs slightly from the class allot discussed below. Optional.
Defaults to a reasonable value, related to avpkt.
The average size of a packet is needed for calculating maxidle,
is the lower cap for packet transmission time calculations -
packets smaller than this value are still deemed to have this
size. Defaults to zero.
When CBQ needs to measure the average idle time, it does so
using an Exponentially Weighted Moving Average which smoothes
out measurements into a moving average. The EWMA LOG determines
how much smoothing occurs. Lower values imply greater sensitiv-
ity. Must be between 0 and 31. Defaults to 5.
A CBQ qdisc does not shape out of its own accord. It only needs to know
certain parameters about the underlying link. Actual shaping is done in
Classes have a host of parameters to configure their operation.
Place of this class within the hierarchy. If attached directly
to a qdisc and not to another class, minor can be omitted.
Like qdiscs, classes can be named. The major number must be
equal to the major number of the qdisc to which it belongs.
Optional, but needed if this class is going to have children.
When dequeuing to the interface, classes are tried for traffic
in a round-robin fashion. Classes with a higher configured qdisc
will generally have more traffic to offer during each round, so
it makes sense to allow it to dequeue more traffic. All weights
under a class are normalized, so only the ratios matter.
Defaults to the configured rate, unless the priority of this
class is maximal, in which case it is set to 1.
Allot specifies how many bytes a qdisc can dequeue during each
round of the process. This parameter is weighted using the
renormalized class weight described above. Silently capped at a
minimum of 3/2 avpkt. Mandatory.
In the round-robin process, classes with the lowest priority
field are tried for packets first. Mandatory.
avpkt See the QDISC section.
This number of packets is used to calculate maxidle so that when
avgidle is at maxidle, this number of average packets can be
burst before avgidle drops to 0. Set it higher to be more toler-
ant of bursts. You can't set maxidle directly, only via this
As mentioned before, CBQ needs to throttle in case of overlimit.
The ideal solution is to do so for exactly the calculated idle
time, and pass 1 packet. However, Unix kernels generally have a
hard time scheduling events shorter than 10ms, so it is better
to throttle for a longer period, and then pass minburst packets
in one go, and then sleep minburst times longer.
The time to wait is called the offtime. Higher values of min-
burst lead to more accurate shaping in the long term, but to
bigger bursts at millisecond timescales. Optional.
If avgidle is below 0, we are overlimits and need to wait until
avgidle will be big enough to send one packet. To prevent a sud-
den burst from shutting down the link for a prolonged period of
time, avgidle is reset to minidle if it gets too low.
Minidle is specified in negative microseconds, so 10 means that
avgidle is capped at -10us. Optional.
Signifies that this class will not borrow bandwidth from its
Means that this class will not borrow bandwidth to its siblings
split major:minor & defmap bitmap[/bitmap]
If consulting filters attached to a class did not give a ver-
dict, CBQ can also classify based on the packet's priority.
There are 16 priorities available, numbered from 0 to 15.
The defmap specifies which priorities this class wants to
receive, specified as a bitmap. The Least Significant Bit corre-
sponds to priority zero. The split parameter tells CBQ at which
class the decision must be made, which should be a (grand)parent
of the class you are adding.
As an example, 'tc class add ... classid 10:1 cbq .. split 10:0
defmap c0' configures class 10:0 to send packets with priorities
6 and 7 to 10:1.
of the average to short bursts. Higher values mean less sensi-
The actual bandwidth of the underlying link may not be known, for exam-
ple in the case of PPoE or PPTP connections which in fact may send over
a pipe, instead of over a physical device. CBQ is quite resilient to
major errors in the configured bandwidth, probably a the cost of
Default kernels rely on coarse timing information for making decisions.
These may make shaping precise in the long term, but inaccurate on sec-
ond long scales.
See tc-cbq-details(8) for hints on how to improve this.
o Sally Floyd and Van Jacobson, "Link-sharing and Resource Manage-
ment Models for Packet Networks", IEEE/ACM Transactions on Net-
working, Vol.3, No.4, 1995
o Sally Floyd, "Notes on CBQ and Guaranteed Service", 1995
o Sally Floyd, "Notes on Class-Based Queueing: Setting Parame-
o Sally Floyd and Michael Speer, "Experimental Results for Class-
Based Queueing", 1998, not published.
Alexey N. Kuznetsov, <firstname.lastname@example.org>. This manpage maintained by
bert hubert <email@example.com>
iproute2 16 December 2001 CBQ(8)
Man Pages Copyright Respective Owners. Site Copyright (C) 1994 - 2013
All Rights Reserved.