Figure 44: Controlling Traffic in a RIP Network with the Incoming Metric
Depending on the RIP network topology and the links between nodes in the network, you might want to control traffic flow through the network to maximize flow across higher-bandwidth links. Figure 44 shows a network with alternate routes between routers A and D.
Figure 44: Controlling Traffic in a RIP Network with the Incoming Metric
In this example, routes to router D are received by router A across both of its RIP-enabled interfaces. Because the route through router B and the route through router C have the same number of hops, both routes are imported into the forwarding table. However, because the T3 link from router B to router D has a higher bandwidth than the T1 link from router C to router D, you want traffic to flow from A through B to D.
To force this flow, you can modify the route metrics as they are imported into router A's routing table. By setting the incoming metric on the interface from router A to router C, you modify the metric on all routes received through that interface. Setting the incoming route metric on router A changes only the routes in router A's routing table, and affects only how router A sends traffic to router D. Router D's route selection is based on its own routing table, which, by default, includes no adjusted metric values.
In the example, router C receives a route advertisement from router D and readvertises the route to router A. When router A receives the route, it applies the incoming metric on the interface. Instead of incrementing the metric by 1 (the default), router A increments it by 3 (the configured incoming metric), giving the route from router A to router D through router C a total path metric of 4. Because the route through router B has a metric of 2, it becomes the preferred route for all traffic from router A to router D.
To modify the incoming metric on all routes learned on the link between router A and router C and force traffic through router B:
Table 40: Modifying the Incoming Metric