What Is EIGRP Stub Routing in Networking?

Imagine you work in a large office building with many departments. Each department has its own mailbox for internal mail, but there is a central mail room that handles all mail coming from outside the building. The mail room knows where every department is located.

The departments do not need to know the full layout of the building or all the other departments — they just need to know how to send mail to the mail room and receive mail from the mail room. EIGRP stub routing works like this. In a computer network, a small remote office router (the stub router) does not need to know about every single network path in the large corporate headquarters.

It only needs to know how to reach the main corporate network, and it only needs to tell the main network about the specific devices in its own office, like printers and computers. The stub feature prevents the remote router from being overwhelmed with routing information it does not need. It also stops the remote router from accidentally becoming a transit point, where traffic from one remote office tries to go through another remote office, which could cause loops and problems.

Think of it like a library card system. A small branch library (the stub) only knows how to return books to the main library and borrow books from the main library. It does not need to know about the inter-library loan system between other branch libraries.

This keeps things simple and efficient. In technical terms, the stub router advertises only certain routes, such as its directly connected networks or a default route, and it ignores more complex routing updates from the core. This reduces memory usage, CPU load, and the risk of routing loops.

It also makes troubleshooting easier because the remote routers have a very simple routing table.

EIGRP (Enhanced Interior Gateway Routing Protocol) stub routing is a Cisco proprietary feature used to optimize routing in hub-and-spoke topologies. In a typical EIGRP network, all routers exchange full routing information with each other, which can lead to large routing tables and high convergence times in large networks. The stub routing feature addresses this by designating certain routers as stub routers.

A stub router is a router that has only one or a few connections to the rest of the network, usually at a remote or branch location. When configured as a stub, the router restricts the types of routes it advertises and, critically, it does not learn routes from its neighbors via the normal EIGRP update process. Instead, the stub router typically advertises only a default route, its directly connected networks, or a summary route to its hub router.

This is achieved using the 'eigrp stub' command under the EIGRP routing process configuration. The command has several options: 'connected' advertises directly connected networks, 'static' advertises static routes, 'summary' advertises summary routes, 'redistributed' advertises redistributed routes, and 'receive-only' makes the router a pure stub that does not advertise any routes. By default, if no option is specified, the stub router advertises connected and summary routes.

The hub router, typically at the corporate headquarters, learns about the stub router's networks and can reach them. The stub router, in turn, learns a default route from the hub router, pointing all unknown traffic toward the hub. This significantly reduces the routing table size on the stub router, lowers CPU and memory usage, and decreases the time needed for the network to converge after a failure.

It also prevents the stub router from being used as a transit router, which is a common cause of routing loops in EIGRP. In a real IT environment, stub routing is deployed extensively in branch offices, retail stores, or remote campuses where the local router only needs to connect local users to the central data center. The configuration is straightforward: on the remote router, under router eigrp [AS number], the network statements are configured for the local networks, and then the 'eigrp stub' command is applied.

On the hub router, no special stub configuration is required, though it should be configured to advertise a default route to the stub routers, often using the 'ip route 0.0.0.0 0.0.

0.0 [next-hop]' and redistributing it into EIGRP or using the 'network 0.0.0.0' command. The stub feature is compatible with both EIGRP named mode and classic mode, and it works with IPv4 and IPv6.

EIGRP stub routing matters because it directly improves network stability, scalability, and performance in real-world IT environments, especially in organizations with many remote locations. In a typical enterprise, there can be hundreds or thousands of branch offices, each with its own router. Without stub routing, every single one of those routers would try to learn the complete routing table, which could include tens of thousands of routes.

This would consume massive amounts of memory and CPU on small, often low-end branch routers, causing them to slow down or crash. It would also increase the time it takes for the network to recover after a failure, because every router would have to recalculate a huge table. By using stub routing, network engineers can dramatically reduce the load on branch routers, meaning they can use cheaper hardware without sacrificing performance.

It also simplifies troubleshooting because the stub router has a very small routing table, making it easy to see if a route is missing or incorrect. Additionally, stub routing prevents routing loops that can occur when a remote router incorrectly advertises a route back into the network, causing packets to bounce around. This is particularly important for networks that rely on EIGRP, as EIGRP does not have the same loop-prevention mechanisms as some other protocols.

From a security perspective, stub routing also limits the information a remote router can learn about the core network, reducing the risk of a compromised branch router exposing sensitive network topology. In cloud and hybrid environments, stub routing is often used to connect virtual private cloud (VPC) networks to on-premises data centers, ensuring that cloud routers do not become overloaded with routes from the enterprise backbone. In summary, stub routing is a fundamental tool for building large, efficient, and reliable networks, and every network professional working with EIGRP must understand it.

EIGRP stub routing appears in several distinct question patterns on Cisco certification exams. The most common type is the direct knowledge question, where the exam asks: 'Which of the following is true about EIGRP stub routing?' and the correct answer is that a stub router does not learn routes from its neighbors.

Another variant asks: 'What routes does an EIGRP stub router advertise by default?' with the correct answer being connected and summary routes. You also see configuration questions where you are given a scenario and must select the correct command to enable stub routing on a remote router, often with options like 'eigrp stub connected' vs 'eigrp stub receive-only'.

Troubleshooting questions are very common. For example: 'A remote office router is configured as an EIGRP stub. Users at the remote office cannot reach the internet. The hub router is advertising a default route.

What is the most likely cause?' The answer might be that the stub router is configured with 'receive-only', which prevents it from learning any routes, including the default route. Another troubleshooting scenario: 'A network engineer notices that the routing table on a branch router has over 10,000 routes.

What can be done to reduce this?' The answer is to configure the branch router as a stub. You also see architecture and design questions: 'In a hub-and-spoke EIGRP network, which feature prevents a spoke router from being used as a transit?'

The answer is stub routing. There are also scenario-based questions with exhibits showing the configuration of hub and spoke routers, and you must determine why the spoke router does not have a route to a specific subnet. In such questions, you need to check if the stub configuration is correct, for example, if the stub router was configured with only the 'connected' option but the target subnet is redistributed from another protocol.

Finally, simulation questions may require you to configure stub routing on a router in a lab environment, using CLI commands and then verifying the configuration using show commands. In these simulations, a common mistake is forgetting to add the 'connected' keyword when the stub router has static routes or redistributed routes that need to be advertised.

EIGRP stub routing is a feature that is both simple to configure and powerful in its effect on network performance. In practice, when you are building a network for a company with many branches, your first instinct should be to enable stub routing on every branch router. This is because the branch routers are often lower-end models (like Cisco ISR 1100 series) that have limited RAM and CPU.

Without stub routing, they would have to store the entire EIGRP topology table, which can easily contain thousands of routes in a large enterprise. Each route uses memory, and each change in the network requires CPU cycles to process updates. By making the branch router a stub, you tell it to only store a few routes: its own local subnets and a default route pointing to the hub.

This reduces memory usage by up to 90% and makes convergence almost instant for the branch. To implement this, you need to access the branch router's CLI. Enter global configuration mode, then go to the EIGRP process.

For example, if your EIGRP autonomous system is 100, type 'router eigrp 100'. Then type 'eigrp stub connected summary'. This tells the branch to advertise its directly connected networks and any manually configured summary routes.

Do not add the 'receive-only' keyword unless you have a specific reason, because then you would need to manually configure a default route on the branch, which adds overhead. On the hub router, you must ensure that a default route is present and is being advertised into EIGRP. The most common way is to configure a static default route on the hub (ip route 0.

0.0.0 0.0.0.0 [next-hop IP]) and then redistribute static into EIGRP using the 'redistribute static' command under the EIGRP process. Alternatively, you can use the 'network 0.0.0.0' command if your hub is the gateway to the internet.

One thing that often goes wrong is that the branch router does not get the default route because the stub configuration is too restrictive. For example, if you use 'eigrp stub receive-only', the branch will not accept any route from the hub, including the default route. Another common issue is that the branch router can reach the hub but not other branches, which is by design — stub routing prevents direct branch-to-branch routing.

If you need branch-to-branch communication, you must route it through the hub. Stub routing also impacts troubleshooting: if a branch user complains they cannot reach a server at another branch, you know the traffic must go through the hub, so you check the hub's routing table. The stub router's simplicity makes it a great diagnostic tool.

In the broader context of network engineering, stub routing is a best practice for any hierarchical network design. It aligns with the principle of keeping the network edge simple and the core intelligent. It also works with other technologies like DMVPN (Dynamic Multipoint VPN) where the hub router acts as the VPN concentrator.

In DMVPN, each spoke router is often configured as an EIGRP stub to prevent the hub from being overloaded with routing updates from hundreds of spokes. By mastering stub routing, you not only pass your exams but also become a better network designer who builds efficient, scalable networks.