Configure BEIGRP dynamic routing protocol

       

Brief introduction of BEIGRP routing protocol

BEIGRP configuration task list

Activate BEIGRP protocol

Configure the sharable percentage of bandwidth

Adjust the arithmetic coefficient of BEIGRP composite distance

Using “offset” to adjust the composite distance of the router

Turn off auto-summary

Customize route summary

Redistribute other routes into the BEIGRP process

Configure other parameters of BEIGRP

The supervision and maintenance of BEIGRP

Examples of BEIGRP configuration

      

This chapter will detail the configuration process of BEIGRP dynamic routing protocol.

Brief introduction of BEIGRP routing protocol

The technology used by BEIGRP is similar to distance vector routing protocol:

l            The router only makes routing decisions with the information provided by directly connected neighbours;

l          The router only provides the routing information it uses to the directly connected neighbors.But, BEIGRP has some main differences with distance vector routing protocol, which entitles it to have more advantages:

l           BEIGRP saves all routes from all neighbours in the topology table, not just the best routes so far;

l          BEIGRP can make query to the neighbors when it is unable to access the destination and no alternative routes are available, so, the convergence speed of BEIGRP can compete with the best link-state protocol.

The introduction of DUAL----Diffused Update Algorithm is vital for BEIGRP’s superiority to other traditional distance vector routing protocol. It always works actively and queries the neighbous when it is unable to access the destination and there is no alternative routes (feasible replacement). As the convergence process is active rather than negative (negatively waiting for the timeout of the routers), so the convergence speed of BEIGRP is very quick.

BEIGRP is a specific routing protocol designed to adapt to the requirements of EIGRP and is directly based on IP. It meets the following requirements of BEIGRP:

l               Dynamically discover new neighbor and the disappearance of old neighbors through “Hello” message;

l               So the transfer of data are all reliable;

l               The transfer protocol permits unicast and multicast data transfer;

l               The transfer protocol itself can adapt to the change of network condition and neighbor responding;

l               BEIGRP can limit the percentage of its occupation of the bandwidth according to the requirements

BEIGRP configuration task list

To complete the configuration of BEIGRP the following tasks are required to be done, among them, the activation of BEIGRP is necessary while others can be decided according to the requirement.

 

 

Activate BEIGRP protocol

In order to create a BEIGRP process, it is required to execute the following commands:  

Step

Command

Purpose

1

router beigrp as-number

Add a BEIGRP process under global configuration mode

2

network network-number network-mask

Add addresses to this BEIGRP process under router configuration mode

 

   After finishing the above configuration, BEIGRP will start to run on all interfaces belonging to this address, discoveres new neighbours through “Hello” and carryes out initial routing interaction through “update”.

   

Configure the sharable percentage of bandwidth

Under default circumstances, BEIGRP can occupy 50% of the bandwidth at most. You may wish to change this default value in order to guarantee the normal interaction of other data, or wishes to adjust the actually usable bandwidth of BEIGRP through the command when the interface is configured with a bandwidth not fit for actual situation.  Under these conditions, you can use the following commands under interface configuration mode:

Command

Purpose

ip beigrp bandwidth-percent percent

Configure the maximum percentage of BEIGRP messages’ occupation of the bandwidth

    

Adjust the arithmetic coefficient of BEIGRP composite distance

Under certain situations, the arithmetic co-efficient of BEIGRP composite distance may need to be adjusted, and finally influences the routing policy of the router. Although the default arithmetic co-efficient of BEIGRP can satisfy most networks, but it may still need to be adjusted under some particular conditions. But this adjust may bring great change to the whole network, so it must be performed by the most experienced engineers.

Use the following command under router configuration mode:

Command

Purpose

metric weights k1 k2 k3 k4 k5

Adjust the arithmetic co-efficient of BEIGRP composite distance

    

Using “offset” to adjust the composite distance of the router

We use offset list to purposely add all incoming and outcoming routes according to the requirement, or the composite distance of certain routes meeting the requirements. The aim of this approach is to finally influence the routing result of the router, and meets our expected result. During the process of configuration, the user can designate access list or application interface in the offset list selectively and according to your requirements, in order to more clearly notify which routes to carry out operations to increase offset. Looking at the following command:

 

Command

Purpose

offset{type number | *} {in | out} access-list-name offset

Apply an offset list

 

    

Turn off auto-summary

The auto-summary of BEIGRP is different with other dynamic routing protocols, and it obeys the following rules:

l    When a BEIGRP process defines several networks, as long as there is at least one sub-net of this networks exists in the BEIGRP topology list, it creates the summary route of the defined network.

l    The established summary route points to interface Null0, and has the distance as the minimum distance of all the sub-nets of the network included in the route. The summary route is also injected into the main IP routing table with the management distance of 5 (unable to be configured)

l    When sending update to the neighbors in different main IP network, the sub-nets summarized by rule1 and rule 2 is cancelled and only a summary route will be sent.

l    Do not perform summary towards the sub-nets of any networks not listed in BEIGRP process definition.

Under certain network situations, you may wish to report every detailed route to the neighbor, and you may use the following commands:  

Command

Purpose

no auto-summary

Turn off auto-summary of the route

   

Customize route summary

When auto-summary cannot meet the requirements, you may configure route summary on each interface running BEIGRP and designate the destination addresses that need performing summary. The interfaces configured for summary will not send the concrete update information of the sub -nets belonging this summary address, while other interfaces are not affected.

Here, the summary operation follows the following rules:

l        After an interface is configured for summary, it creates the summary route of the defined network as long as this network has at least one sub-net in BEIGRP topology list;

l    The summary route points to interface Null0, and has the distance as the minimum distance of all the sub-nets of the network included in the route. The summary route is also injected into the main IP routing table with the management distance of 5 (unable to be configured)

l           When sending route update on the interface configured with summary range, the concrete route belonging to the summary address will be canceled. The update to other interfaces are not affected.

Command

Purpose

ip beigrp summary-address as-number address mask

Configure route summary on an interface.

 

Redistribute other routes into the BEIGRP process

The redistribute  operation follows the below rules:

l  It isn't have to configure the command "default-metric" when redistribute the static routes and the connected routes. The related parameter(such as: bandwidth, delay, reliability , load and MTU ) is attained from the related interface.

l  It isn't necessary  to configure the command "default-metric" when redistribute  the routes of other beigrp process. The related parameter is attained from the BEIGRP process redistributed.

l     It is necessary to configure the command "default-metric" when redistribute the routes of others protocol  (such as: rip, ospf). The related parameter is validated by the configuration  of "default-metric". If we redistribute the routes of these types without the command "default-metric", the redistribution doesn't work.

In a router running the BEIGRP protocol and the RIP protocol, the following commands must be configured when we need obtain the routes from RIP protocol to BEIGRP protocol.  

default-metric bandwidth delay reliability loading mtu

configure the default parameter of redistribute

redistribute protocol [process] [route-map name]

redistribute the routes to BEIGRP protocol.

Configure other parameters of BEIGRP

In order to adapt to different network environments, and to make BEIGRP be more effectively and fully functions, we may need to adjust the following parameters:

Adjust the time interval of BEIGRP to send “hello” messages and the timeout death time of the neighbours

Turn off split-horizon

    

Adjust the time interval of BEIGRP to send "hello" messages and the timeout death time of the neighbors

BEIGRP hello protocol archieves 3 objectives to enable correct BEIGRP operation:

l        It discovers accessible new neighbors. The discovery is automatic and requires  no manual configuration;

l        It checks neighbors’ configuration and only permits communication with the  neighbours configured with compatible mode.

l        It continues to maintain the availability of the neighbors and detects the disappearance of the neighbors.

The router sends “hello” multicast packet on all interfaces running BEIGRP. All routers support BEIGRP receive these multicast groups, so that it can discover all neighbours.

 “Hello” protocol uses two timers to detect the disappearance of the neighbours: hello interval defines the frequency of sending BEIGRP hello messages on the interface of the router, while hold timer defines the interval of time the router has to wait for the communication data from the designated neighbor before the declaration of the neighbour’s death. We ordered that every time it receives BEIGRP packet from the neighbour router, it resets the hold timer.

Different network type or network bandwidth will use different default value of hello timer:

Interface type encapsulation

 

Hello timer (second)

Hold timer (second)

LAN interface

Any

5

15

WAN interface

HDLC or PPP

5

15

NBMA interface, bandwidth<=T1

60

180

NBMA interface, bandwidth>T1

5

15

The point-to-point sub-interface of NBMA interface

5

15

 

The difference of the default value of the timer in Hello protocol may induce the result that the BEIGRP neighbours connected to different IP sub-network use different hello and hold timer. To resolve the problem, the hello packet of every router designates its own hold timer, every BEIGRP router uses neighbour’s the designated hold timer of the hello group to decide the timeout of this neighbour. Here, it can enable the appearance of different neighbour error detection timers in the different stands of the same WAN nephogram. But under some particular situation, the default value of the timer cannot be met, so if you want to adjust the time interval of sending hello messages, use the following command:

Command

Purpose

ip beigrp hello-interval seconds

Adjust the time interval of sending hello message from this interface

  

If you wish to adjust the timeout timer of the neighbour, use the following command:

Command

Purpose

ip beigrp hold-time seconds

Adjust the timeout death time of the neighbor

Turn off split-horizon

Commonly, we wish to use split-horizon. It will prevent the routing information from one interface to be broadcasted back to the same interface, so as to avoid route loop. But under certain circumstances, this is not the optimized choice, and then we can use the following command to disable split-horizon: 

Command

Purpose

no ip beigrp split-horizon

Turn off horizontal split

 

The supervision and maintenance of BEIGRP

To clear the neighbourship with all neighbours, use the following command:

Command

Purpose

clear ip beigrp neighbors [as-number | interface]

To clear the neighborship with all neighbours

 

In order to show various statistics information of BEIGRP, execute the following commands:

Command

Purpose

show ip beigrp interfaces [interface] [as-number]

show interface information

show ip beigrp neighbors [as-number | interface]

show neighbor information

show ip beigrp topology [as-number | all-link | summary | active]

show topology information

 

Examples of BEIGRP configuration

An example of route summary

In the following example, the interface ethernet1/1 is configured to send a summary route to address 10.0.0.0/8, and all subnets routes belonging to this address will not be advertised to the neighbors from the interface.  Meanwhile, we turn off the auto-summary of BEIGRP process.

interface Ethernet 1/1

ip beigrp summary-address 1 10.0.0.0 255.0.0.0

!

router beigrp 1

network 172.16.0.0 255.255.0.0

    no auto-summary