IGRP Features and Operation

IGRP Features and Operation
IGRP sends out periodic broadcasts of its entire routing table. Upon initialization, IGRP
broadcasts a request out all IGRP-enabled interfaces. IGRP then performs a check on the
received updates to validate that the source address of the update belongs to the same subnet
that the update was received on. Each router then uses the learned routes to determine the best
routes to every destination network.
Like RIPv1, IGRP is a classful routing protocol. IGRP does not send out subnet masks with its
routing updates. This means that IGRP cannot be used on a network that has employed VLSM.
IGRP employs the use of autonomous systems (ASs). If you remember, an AS is a collection
of routing domains under the same administrative control. An IGRP AS is actually a routing
domain and not a true AS. An AS can have multiple instances of IGRP and other routing protocols
running concurrently. An IGRP AS is a set of routers running the IGRP routing protocol
with the same AS number.
Allowing multiple IGRP ASs to exist under a single AS means an administrator can better
segment an internetwork. The administrator will be able to create an IGRP AS for each routing
domain, which means better control of communications throughout the internetwork and
between routing domains through redistribution.
IGRP recognizes three types of routes within its updates:
Interior
Networks directly connected to a router interface.
System
Routes advertised by other IGRP neighbors within the same IGRP AS.

Exterior
Routes learned via IGRP from a different IGRP AS, which provides information used
by the router to set the gateway of last resort. The
gateway of last resort
is the path a packet will
take if a specific route isn’t found on the router.
IGRP has other features that are briefly described in Table 4.1. Most of these features were
added to make IGRP more stable; a few were created to deal with routing updates and make
network convergence faster.