EIGRP

What is EIGRP?

In a well-architected network, Enhanced Interior Gateway Routing Protocol (EIGRP) offers excellent scalability and achieves rapid convergence with minimal network traffic. Key advantages of EIGRP include:

• Minimal resource consumption during normal operations, as only hello packets are sent in a stable environment.

• Upon changes, only the necessary updates to the routing table are communicated, avoiding the need to send the entire routing table and thus reducing the protocol's impact on network performance.

• Quick convergence times in response to topology changes, with some scenarios allowing for nearly instantaneous convergence.

EIGRP is an advanced distance vector protocol that uses the Diffused Update Algorithm (DUAL) to determine the shortest path to a destination within the network.

Quick Definition

A typical distance vector protocol records the distance (such as hop count) and the next hop when determining the best path to a destination. For example, in the Figure below, all routers run RIP. Router Two selects the path to Network A by comparing hop counts. Since the route via Router One has two hops and the one through Router Three has three, Router Two chooses the path through Router One, discarding the route through Router Three. If the link between Router One and Network A fails, Router Two loses access until the route times out after three update periods (90 seconds), and Router Three re-advertises it (every 30 seconds in RIP). Without hold-down time, switching to Router Three can take 90-120 seconds.

Unlike RIP, EIGRP does not rely on full periodic updates for re-convergence. Instead, it maintains a topology table with data from its neighbors and selects a loop-free route from this table. If no alternative is found, it queries its neighbors. Router Two stores information from both Routers One and Three, using Router One as the primary route (successor) and Router Three as a backup (feasible successor). If the link through Router One fails, Router Two immediately switches to the path via Router Three.

To achieve this, EIGRP provides:

• Efficient updates only when necessary, supported by neighbor discovery and maintenance.

• A mechanism to identify loop-free paths.

• A process for removing bad routes from all routers' topology tables.

• A method for querying neighbors to find alternate paths to lost destinations.

EIGRP Neighbor Discovery and Hello Packet Mechanism

EIGRP uses non-periodic incremental updates to share routing information, only sending updates when changes occur in the network. However, to detect when a path becomes unavailable, EIGRP establishes neighbor relationships by exchanging hello packets on a shared network. Hello packets are sent every 5 seconds on high-bandwidth links and every 60 seconds on low-bandwidth multipoint links. The rate at which these packets are sent is called the hello interval, which can be adjusted per interface. The hold time, usually three times the hello interval, determines how long a router waits before considering a neighbor down. Both the hello interval and hold time can be modified manually.

Reported Distance, Feasible Distance, and Feasible Successors in EIGRP

Feasible distance is the best metric to a destination network, including the metric to the neighbor that advertises the path. Reported distance represents the total metric along the path to a destination as indicated by an upstream neighbor. A feasible successor is a path where the reported distance is less than the feasible distance (the current best path).

For example, Router One identifies two routes to Network A: one through Router Three and another through Router Four. The route via Router Four has a cost of 46277376 and a reported distance of 307200, while the route through Router Three has a cost of 20307200 and the same reported distance. EIGRP selects the route through Router Three as the best path, setting its metric as the feasible distance. Since the reported distance through Router Four is less than the feasible distance, it is classified as a feasible successor.

If the link between Routers One and Three fails, Router One checks its known paths to Network A and finds the feasible successor through Router Four. It adopts this route, updating the feasible distance accordingly. The network converges quickly, with only updates to downstream neighbors being transmitted.

In a more complex scenario, Router One has two routes to Network A: one through Router Two (metric 46789376) and another through Router Three (metric 20307200). Router One selects the lower metric from Router Three as its feasible distance. The reported distance from Router Two is 46277376, which exceeds the feasible distance, thus disqualifying it as a feasible successor. At this point, Router One's topology table shows only the entry through Router Three.

If the link to Router Three fails, Router One queries its neighbors for alternate routes. Since Router Two has a route to Network A, it responds. Router One then accepts this route through Router Two, as it no longer has access to the path through Router Three.

Ensuring Loop-Free Path Discovery

EIGRP utilizes the concepts of feasible distance, reported distance, and feasible successor to assess the validity of paths and prevent loops. In a scenario where Router Three examines routes to Network A, with split horizon disabled, it identifies three potential routes: one through Router Four, another via Router Two, and the last through Router One.

If Router Three accepts all these routes, it risks creating a routing loop. For example, Router Three may think it can reach Network A through Router Two, but this path actually loops back through Router Three. If the link between Routers Four and Three fails, Router Three could mistakenly believe it has alternative routes to Network A, yet the rules for determining feasible successors would prevent these paths from being utilized.

Examining the metrics reveals:

• Total metric to Network A via Router Four: 20281600

• Total metric to Network A via Router Two: 47019776

• Total metric to Network A via Router One: 47019776

Since the path through Router Four has the lowest metric, Router Three installs this route in its forwarding table, designating 20281600 as its feasible distance. It then computes the reported distances to Network A through Routers Two and One, both yielding 47019776. As these values exceed the feasible distance, Router Three does not recognize either as a feasible successor.

If the link between Routers Three and Four fails, Router Three queries its neighbors for alternative routes. Router Two, upon receiving this query, checks its topology table for feasible successors and finds only Router One's entry, which does not qualify as a feasible successor since its reported distance equals the last known metric through Router Three. Consequently, Router Two marks this route as unreachable and queries Router One.

Router Three also queries Router One, which finds its only path to Network A through Router Two, again with an insufficient reported distance. As a result, Router One marks the route as unreachable and queries Router Two.

This initiates the first level of queries. Router Three has sought routes from its neighbors, and Routers One and Two have marked the path as unreachable. When Router Two receives Router One's query, it confirms the destination is unreachable, responding accordingly. Router One, upon receiving Router Two's response, also affirms that Network A is unreachable.

At this point, both Routers One and Two conclude that Network A is unreachable and relay this information back to Router Three. The network has converged, and all routes revert to a passive state.

The Advantages of EIGRP in Real-World Scenarios

EIGRP is highly effective in large, complex networks due to its ability to send updates only when necessary, reducing traffic and ensuring efficiency. It quickly discovers new routes when part of the network fails, keeping operations running smoothly.

By supporting multiple paths, EIGRP ensures that if one route fails, another can take over without interruption. This redundancy is critical for maintaining high availability, ensuring continuous network uptime.

In large enterprises, EIGRP’s scalability and rapid convergence minimize downtime and enhance data routing efficiency. In VPN deployments, EIGRP handles multiple paths and provides fast updates, making it ideal for dynamic, multipoint setups that need reliable data transmission across multiple locations. Its advanced features and design make EIGRP an excellent choice for optimizing performance and stability in complex network environments.

Conclusion

EIGRP stands out as an advanced and adaptable routing protocol that excels in complex, large-scale networks. Its efficiency in handling dynamic environments, quick convergence, and integration with other protocols like OSPF and BGP make it a valuable asset for enterprises. EIGRP’s ability to maintain high availability through redundancy, scale across multiple sites, and seamlessly exchange routing information ensures it remains a powerful solution in modern networking. For those looking to work in enterprise environments, understanding EIGRP is essential due to its continued relevance and widespread use.