A routing protocol is a method that routers use to share information with each other about networks they are connected to.
We have seen in previous posts on this website (Creating a Static Route and Default Gateway of Last Resort) how we make routes ourselves and how to teach a router how to send packets to networks that are not directly connected to it. Using static routes can be useful in some cases however in large networks it would take us a long time to manually teach every router about every network that we have.
The second problem with creating static routes is we are unable to monitor the links and whether they are operational or not. So a router will always send a packet along the path we have told it without checking if the path is the best option or even a valid option.
For these reasons a routing protocol is a better solution.
Once we enable the same routing protocol on multiple routers, they are able to communicate with each other and share information about all the networks they are directly connected to and they are able to monitor if a network goes down due to a link failure. Since all routers need to run the same routing protocol to communicate with each other, you can almost think of this as a language.
If a router receives information about a network from multiple routers, it is able to discover there is more than one path to that network and then looks at a value called the ‘Cost’ or ‘Metric’ to work out which path is the best to use.
Each routing protocol has a different approach in sending information and what type of information is sent. Routing protocols such as RIP send information at set periodic intervals and only report how many hops it is to the destination network. In comparison, routing protocols like OSPF and EIGRP send information only when something has changed or there is something new to report, and they send information about the link speeds, their reliability and delay. This makes routing protocols like OSPF and EIGRP far more reliable than RIP. For this reason each routing protocol also has a ‘trust’ factor known as the administrative distance which the router uses to determine which routing protocol it will trust when making a decision as to which path is the best.
Routing protocols are often separated into “Distance Vector” or “Link-State”.
Distance Vector routing protocols, such as RIP and IGRP, derive their cost of the route based on how many devices are along the path – Hop Count.
Link-State protocols, such as OSPF and IS-IS, also look at information about the links to identify which path will be the fastest, even if this means sending the packet the longest distance.
EIGRP is often placed under distance vector routing protocols but since it also has a lot of the traits of a link-state protocol it falls into its own category – “Hybrid”