BANDWIDTH MANAGER
Introduction
Bandwidth manager is used to control the bandwidth. Maximum Bangladeshi ISPs are using software bandwidth manager. If we have less than 5 Mbps bandwidth, you can control it easily by a software bandwidth manager.
There are a lot of way and utility, left to control your traffic; it is not even possible to write all names in this instance. We will prioritize only known and wide use software bandwidth controllers. You may find many operating systems with built-in bandwidth controller facility. There are some customized software based bandwidth control systems such as CBQ, HTB etc. available in market. We will describe two bandwidth control method CBQ and HTB in this chapter.
CBQ
Hierarchical Class-Based Queuing is a queuing algorithm to manage Bandwidth. Through user-definable class definitions, incoming packet traffic is divided into classes. These divisions might fall along the lines of traffic from a given interface, associated with a particular application, intended for a particular network or device destination, and all traffic of a specific priority classification.
Many Linux OS are using CBQ to control traffic. Many commercial bandwidth-management package vendors use CBQ as backend tool. It is most known and used bandwidth management technique nowadays. It became very popular for free distribution. There are some GNU Licensed tools available to support CBQ. It is an open technology from University of California, Berkeley.
HTB
Hierarchical Token Bucket. HTB is meant to be a more understandable, intuitive and faster replacement for the CBQ qdisc in Linux. Both CBQ and HTB help you to control the use of the outbound bandwidth on a given link. Both allow you to use one physical link to simulate several slower links and to send different kinds of traffic on different simulated links. In both cases, you have to specify how to divide the physical link into simulated links and how to decide which simulated link to use fora given packet to be sent.
Configuring the bandwidth manager
It is important to choose a mode for bandwidth management in Linux. Linux can work in both route and bridge mode to provide bandwidth management facility.
Routing
Routing mode is the common approach for bandwidth management in Bangladesh. In routing mode, Bandwidth Manager works as a complete router. It contains 1P address into all interfaces and each individual route separately.
Routing mode is comparatively easy to deploy in any OS. But it is a bit hard to maintain all routes. But if we use routing mode, you will be able to able to keep maximum routes into your bandwidth manager. It is easy to manage, monitor, and debug. In routing mode, if your bandwidth manager goes down the total process will go down.
Bridging
Bride mode is sometime called “IP Transparent Mode”. In bride mode. Band’.', id Manager works as a HUB/Switch. Bridge type Manager may not contain IP addresses into all interfaces. It may not contain all routes separately.
Bridge is comparatively a bit hard to deploy but you need to go through a less hassle for route management. But it is pretty weak for distributed routing and policy management. You need to recompile the kernel again to enable bride mode.
If you use bridge mode, router will maintain all routes. In bride mode, if Bandwidth Manager goes down, there will be no effect to the network (you can just plug it out), only the bandwidth will be unmanaged. Backing up a bridge based bandwidth controller is easier than a routed one.
We will use bridging mode for bandwidth control.
Configuration a linux pc as a bridging mode
Software required
# rpm –ivh bridge*
Note: bridge-utils-0.9.3-8 and bridge-utils-devel-0.9.3-8 file will be found.
# brctl addbr br0
# brctl addif br0 eth0 (or eth1)
# ifconfig eth0 (or eth1) 0.0.0.0 up
# ifconfig br0 192.168.110.0 up
# ifconfig | less
Note: when we use a PC as a switch or bridge, then the ip address of the LAN interfaces of that PC must set to 0.0.0.0
Configuration of CBQ
This description is meant to simplify setup and management of relatively simple CBQ-based traffic control on Linux. CBQ is to be implemented to the NIC card.
Software required
cbq.init v0.7.2
# cp cbq.init v0.7.2 /etc/init.d/cbq.init
File name
Every traffic class must be described by a file in the $CBQ_PATH directory
(/etc/sysconfig/cbq by default) – one file per class.
The config file names must obey mandatory format: cbq-<clsid>.<name> where
<clsid> is two-byte hexadecimal number in range <0002-FFFF> (which in fact
is a CBQ class ID) and <name> is the name of the class — anything to help
you distinguish the configuration files. For small amount of classes it is
often possible (and convenient) to let <clsid> resemble bandwidth of the class.
Example of valid config name: cbq-1280.My_first_shaper.
The configuration file may contain the following parameters:
Device parameters
DEVICE=<ifname>, <bandwidth> [<weight>] mandatory
DEVICE=eth0, 10Mbit, 1Mbit
| Parameters | Description |
| <ifname> | It is the name of the interface you want to control traffic on, e.g. eth0. |
| <bandwidth> | It is the physical bandwidth of the device, e.g. for ethernet 10Mbit or 100Mbit, for arcnet 2Mbit. |
| <weight> | It is tuning parameter that should be proportional to <bandwidth>. As a rule of thumb: <weight> = <bandwidth> / 10 |
Note: When you have more classes on one interface, it is enough to specify <bandwidth> [and <weight>] only once, therefore in other files you only
need to set DEVICE=<ifname>.
Class parameters
| Class parameters | Description |
| RATE | RATE=<speed>(mandatory). Bandwidth allocated to the class. Traffic going through the class is shaped to conform to specified rate. We can use Kbit, Mbit or bps, Kbps and Mbps as suffices. If we don’t specify any unit, bits/sec are used. Also note that “bps” means “bytes per second”, not bits. |
| WEIGHT | WEIGHT=<speed>(mandatory). Tuning parameter that should be proportional to RATE. As a rule of thumb, use WEIGHT = RATE / 10. |
| PRIO | PRIO=<1-8>(optional, default 5). Priority of class traffic, the higher the number, the lesser the priority. Priority of 5 is just fine. |
Table: Class parameters
Filter parameters
RULE=[[saddr[/prefix]][:port[/mask]],][daddr[/prefix]][:port[/mask]]
These parameters make up “u32″ filter rules that select traffic for each of the classes. We can use multiple RULE fields per config.
| Example | |
| RULE=10.1.1.0/24:80 | selects traffic going to port 80 in network 10.1.1.0. |
| RULE=10.2.2.5 | selects traffic going to any port on single host 10.2.2.5. |
| RULE=10.2.2.5:20/0xfffe | selects traffic going to ports 20 and 21 on host 10.2.2.5 |
| RULE=:25,10.2.2.128/26:5000 | selects traffic going from anywhere on port 50 to port 5000 in network 10.2.2.128. |
| RULE=10.5.5.5:80 | selects traffic going from port 80 of single host 10.5.5.5 |
MARK=<mark>
These parameters make up “fw” filter rules that select traffic for each of the classes accoring to firewall “mark”. Mark is decimal number packets are tagged with if firewall rules say so. You can use multiple MARK fields per config.
Example configuration
| Example Filename | Contents |
| cbq-1003.root | DEVICE=eth1,10Mbit,1Mbit RATE=28Kbit WEIGHT=2Kbit PRIO=5 RULE=192.168.1.1 MARK=7 |
| cbq-2008.root | DEVICE=eth0,10Mbit,1Mbit RATE=128Kbit WEIGHT=10Kbit PRIO=5 RULE=192.168.1.1 MARK=7 |
Table: Example configuration
Note
For downloading MARK is optional. For uploading MARK is mandatory.
To start cbq
# /etc/init.d/cbq.init start
To stop cbq
# /etc/init.d/cbq.init stop
Configuration of HTB
Software required
cbq.init v0.7.2
# cp cbq.init v0.7.2 /etc/init.d/cbq.init
File name
Every traffic class must be described by a file in the $HTB_PATH directory
(/etc/sysconfig/htb by default) – one file per class.
Class options belong to files with names matching this expression:
$HTB_PATH/<ifname>-<clsid>(:<clsid>)*<description>
<clsid> is class ID which is hexadecimal number in range 0×2-0xFFFF, without
the “0x” prefix. If a colon-delimited list of class IDs is specified, the
last <clsid> in the list represents ID of the class in the config file.
<clsid> preceding the last <clsid> is class ID of the parent class. To keep
ordering so that parent classes are always created before their children, it
is recommended to include full <clsid> path from root class to the leaf one.
<description> is (almost) arbitrary string where you can put symbolic
class names for better readability.
Examples of valid names:
eth0-2 root class with ID 2, on device eth0
eth0-2:3 child class with ID 3 and parent 2, on device eth0
eth0-2:3:4 child class with ID 4 and parent 3, on device eth0
eth1-2.root root class with ID 2, on device eth1
HTB qdisc parameters
| Parameters | Description |
| DEFAULT | DEFAULT=<clsid>(optional, default 0) Example DEFAULT=30 <dclsid> is ID of the default class where UNCLASSIFIED traffic goes. Unlike HTB qdisc, HTB.init uses 0 as default class ID, which is internal FIFO queue that will pass packets along at FULL speed. |
| R2Q | R2Q=<number>(optional, default 10) Example R2Q=100 This allows you to set coefficient for computing DRR (Deficit Round Robin) quanta. The default value of 10 is good for rates from 5-500kbps and should be increased for higher rates. |
Table HTB qdisc parameters
HTB class parameters
| Parameters | Description |
| RATE | RATE=<speed> or prate or pceil Example RATE=5Mbit Bandwidth allocated to the class. Traffic going through the class is shaped to conform to specified rate. You can use Kbit, Mbit or bps, Kbps and Mbps as suffices. If you don’t specify any unit, bits/sec are used. Also note that “bps” means “bytes per second”, not bits. |
| CEIL | CEIL=<speed>|prate|pceil(optional, default $RATE). Example CEIL=6MBit. |
| BURST | BURST=<bytes> (optional, default computed. Example BURST=10Kb. |
| PRIO | PRIO=<number>(optional, default 0) Example PRIO=5 Priority of class traffic. The higher the number, the lesser the priority. |
| LEAF | LEAF=none or sfq or pfifo or bfifo (optional, default “none”). LEAF tells the script to attach specified leaf queueing discipline to HTB class. By default, no leaf qdisc is used. |
Filter parameters
Same as CBQ described previously in this chapter. Example configuration
| Device | Example Filename | Contents |
| eth0 | htb-2.110.root | RATE=512Kbit CEIL=128Kbit BURST=15Kbit LEAF=sfq MARK=4 RULE=192.168.110.3/32 |
| eth1 | cbq-2008.root | DEVICE=eth0,10Mbit,1Mbit RATE=128Kbit WEIGHT=10Kbit PRIO=5 RULE=192.168.1.1 MARK=7 |
Table: Example configuration
Note
For downloading MARK is optional. For uploading MARK is mandatory.
To start cbq
# /etc/init.d/htb.init start
To stop cbq
# /etc/init.d/htb.init stop
In this way can control the bandwidth.
Hiç yorum yok:
Yorum Gönder