Instruction Manual

The lower the metric is, the better the route. Make sure that students understand that the

default metrics for IGRP are bandwidth and delay. The other metrics can be used but are not

used by default. Allow the students to configure IGRP on a mesh network and adjust the

metrics to see how the routing table is changed. Students should also be instructed to trace

the path to a network before and after the metric changes to verify the different path selection.

7.3.3 IGRP routes

IGRP advertises three types of routes:

• Interior

• System

• Exterior

Interior routes are defined as routes between subnets that are connected to the same router

interface. System routes are routes within the same autonomous system. These routes are

derived from directly connected networks and through routes learned from other IGRP routers.

System routes do not contain subnet information. Exterior routes are routes between

autonomous systems. A gateway of last resort can be used to transfer information to a

destination outside of a local autonomous system.

Describe the graphic included in the TI in depth. Explain the concepts of interior routes and

multiple subnets on the same router interface. Autonomous systems should also be explained.

7.3.4 IGRP stability features

Features designed to enhance the stability of IGRP consist of holddowns, spit horizons, and

poison-reverse updates. Holddowns are used to prevent regular update messages from

reinstating a route that is down. This is done through the lack of regularly scheduled update

messages. If a router does not receive an update about a particular route, it marks that route

as possibly down. Split horizons are designed to prevent routing loops with the rule that

routing information is not sent back in the direction from which it was learned. This prevents

routing loops between adjacent routers. Poison-reverse updates are necessary to avoid larger

routing loops. An increase in metric may indicate a routing loop, so poison-reverse updates

are sent to place the route with the increasing metric in holddown. IGRP sends out poison-

reverse updates when the route metric has increased by a factor of 1.1 or more.

The timers associated with IGRP include update, invalid, hold-down, and flush timers. The

update timer indicates how often routing updates will be sent, the default for IGRP is 90

seconds. The invalid timer is the amount of time that IGRP will wait before it declares a route

invalid. The default for IGRP is 270 seconds, which is 3 times the update period. The hold-

down variable specifies the holddown period. During this period the information about better

routes is suppressed, even though the route in holddown is marked as inaccessible and

advertised as unreachable. When the holddown time has expired, routes advertised by other

routers are accepted. The default holddown time is greater than three times the update time.

The flush timer indicates the amount of time that a route should remain in the routing table

before it is flushed. This time should be at least as long as the holddown and invalid times

combined. This will allow the proper holddown phase, otherwise the route may be flushed and

new routes may be accepted prematurely. The default flush timer is seven times the update

time. The show ip protocol command should be used to view the timers and then timers

should be changed and viewed again.