Route redistribution is a technique that allows routes learned by one routing protocol or algorithm to be advertised into another routing protocol or algorithm. For the simplicity and ease of management of an internetwork, it is always preferred to use a single routing protocol or algorithm. However, this is not always possible, because each internetwork has its own requirements, limitations, and conditions. For example, consider an internetwork that consists of 100 of routers of different vendors such as Cisco and Juniper. You cannot use the EIGRP routing protocol, because it is a Cisco-proprietary protocol and does not support routers of other vendors such as Juniper. So, we want to say that the multi-protocol implementation is common in the multi-vendor environments.
Now, the question is that how the routing information are shared from one routing protocol or algorithm to another routing protocol or algorithm. The answer is Router Redistribution. Routes learned by the Route Redistribution technique are marked as external routes in the routing table. The external routes have less preference over the locally-originated routes.
In order to redistribute routes from one routing protocol to another routing protocol, there must be at least one redistribution point (device). The redistribution device will run both routing protocols. For example, if you want to redistribute the routes learned by the RIP routing protocol into the EIGRP routing protocol, at least one interface of the redistribution device should be included in the RIP routing domain and at least one another interface should be included in the EIGRP routing domain.
In the following figure, you can see that Router2 acts as a distribution device between RIP and EIGRP routing domains.
Apart from redistributing routes from one routing protocol to another routing protocol, it is also possible to redistribute routes between the same routing protocols. For example, consider an internetwork that consists of two separate OSPF domains: one is configured with 100 process ID and another one is configured with 200 process ID. In addition, static routes and connected interfaces can also be redistributed into a routing protocol such as RIP and OSPF.
Note: Keep in mind, routes that are stored in a topology database (topology table) are not redistributed.
One of the most important factor that needs to be consider while performing route redistribution is Metric. Each routing protocol uses a unique metric to calculate the best path to a destination network. For example, the RIP protocol uses the hop count metric while the EIGRP protocol uses the composite (bandwidth, load, delay, and reliability) metric. The metric value of an injecting protocol (the protocol of which you want to redistribute routes) should be defined in such a manner that it can be understood by the receiving protocol (the protocol to which you want to redistribute routes).
Answer the following question
Which route should be preferred the best route if a router runs more than one routing protocol, such as RIP and EIGRP, and learns a route to the same destination using both routing protocols?
The answer is pretty straightforward: the route that has the best metric value would be considered as the best route. No, it’s not the right answer, because the RIP (hop count) and EIGRP (composite) both use its own metric type to determine the best route. Routes having different metric types cannot be compared. Then, how the best route would be calculated? Here, the Administrative Distance (AD) values will play an important role. Routing protocol having the lower AD value would be preferred in this type of scenarios.