To achieve basic connectivity between all RIP hosts in a RIP network, you enable RIP on every interface that is expected to transmit and receive RIP traffic. To enable RIP on an interface, you define RIP groups, which are logical groupings of interfaces, and add interfaces to the groups. Additionally, you must configure a routing policy to export directly connected routes and routes learned through RIP routing exchanges.
To tune a RIP network and control traffic flowing through the network, you increase or decrease the cost of the paths through the network. RIP provides two ways to modify the path cost: an incoming metric and an outgoing metric, which are each set to 1 by default. These metrics are attributes that manually specify the cost of any route advertised through a host. By increasing or decreasing the metrics—and thus the cost—of links throughout the network, you can control packet transmission across the network.
The incoming metric modifies the cost of an individual segment when a route across the segment is imported into the routing table. For example, if you set the incoming metric on the segment to 3, the individual segment cost along the link is changed from 1 to 3. The increased cost affects all route calculations through that link. Other routes that were previously excluded because of a high hop count might now be selected into the router's forwarding table.
The outgoing metric modifies the path cost for all the routes advertised out a particular interface. Unlike the incoming metric, the outgoing metric modifies the routes that other routers are learning and thereby controls the way they send traffic.
RIPv2 provides authentication support so that RIP links can require authentication keys (passwords) before they become active. These authentication keys can be specified in either plain-text or MD5 form. Authentication provides an additional layer of security on the network beyond the other security features.
This type of authentication is not supported on RIPv1 networks.