When most people hear the words “routing protocol,” their eyes glaze over. But if you’re in the IT game, especially on the networking side, understanding these protocols is absolutely critical. And one of the big names you’ll run into is EIGRP: the Enhanced Interior Gateway Routing Protocol.
Now, if you’ve ever worked with Cisco gear (or plan to) EIGRP is going to come up a lot. It’s proprietary, it’s powerful, and it’s packed with features that help networks run efficiently.
So, what is EIGRP really? How does it work? And why should you care?
What Is EIGRP?
EIGRP stands for Enhanced Interior Gateway Routing Protocol. It was developed by Cisco as an improvement over its older protocol, IGRP (Interior Gateway Routing Protocol). While IGRP had some strengths, it wasn’t designed to scale well as networks became larger and more complex. So Cisco decided to go back to the drawing board and build something better, something that could handle the demands of modern networks.
EIGRP is classified as an advanced distance-vector routing protocol, though many refer to it as a hybrid protocol because it blends features of both distance-vector and link-state protocols. That gives it a unique sweet spot in terms of speed, scalability, and efficiency.
It was Cisco-proprietary for many years, which meant you could only use it on Cisco devices. But in 2013, Cisco released EIGRP to the IETF as an open standard )at least partially) making it more accessible.
Why Do We Use EIGRP?
Why use EIGRP instead of something like RIP or OSPF?
Here’s the deal:
- Faster convergence: EIGRP reacts quickly to network changes.
- More efficient routing decisions: It uses metrics based on bandwidth, delay, reliability, and load.
- Support for unequal-cost load balancing: Most protocols support equal-cost load balancing, but EIGRP lets you use multiple paths with different metrics.
- Scalability: It works well in both small and large enterprise networks.
- Flexibility: It supports multiple network layer protocols like IPv4, IPv6, IPX, and AppleTalk (although most people only care about IPv4 and IPv6 now).
If you’re working in a Cisco-heavy environment, EIGRP is often a go-to protocol for dynamic routing.
EIGRP Metrics: How Routes Are Calculated
EIGRP doesn’t just look at hop count like RIP does. It uses a composite metric, which means it takes multiple factors into account to choose the best path.
These factors include:
- Bandwidth: The slowest link along the route.
- Delay: The cumulative delay across all links in the path.
- Reliability: A value between 0 and 255, with 255 being 100% reliable.
- Load: How busy a link is, also on a scale from 0 to 255.
- MTU (Maximum Transmission Unit): Used in tie-breaking, not the metric itself.
By default, EIGRP only uses bandwidth and delay in its metric calculation, but the others are available.
The formula for EIGRP’s metric (simplified) is:
- Metric = (10^7 / lowest bandwidth) + cumulative delay
And that metric determines the “best” path.
Key Concepts Behind EIGRP
Let’s pause and walk through some of the terms and mechanisms that make EIGRP tick.
DUAL – Diffusing Update Algorithm
DUAL is the heart of EIGRP. It ensures loop-free and backup routes without depending on slow convergence.
Here’s what DUAL does:
- Determines the best route to a destination (called the successor).
- Identifies backup routes (called feasible successors) that can be used immediately if the main route fails.
- Maintains loop-free routing by checking the feasibility condition.
Successor and Feasible Successor
- Successor: The best, primary route.
- Feasible Successor: A backup route that’s loop-free and meets the feasibility condition.
If a route’s reported distance (RD) from a neighbor is less than the local feasible distance (FD), it qualifies as a feasible successor.
This allows for instant failover…no recalculating needed.
Neighbor Relationships
EIGRP routers form neighbor relationships by sending Hello packets. Once neighbors are established, they exchange routing information.
Default Hello intervals:
- 5 seconds on high-speed links (e.g., Ethernet)
- 60 seconds on low-speed links (e.g., T1)
If a router doesn’t hear from a neighbor within the Hold Time (usually 3x the Hello interval), it assumes the neighbor is down and recalculates routes.
Topology Table
Each EIGRP router maintains a topology table. This is a record of all learned routes (even ones not currently being used).
It includes:
- All available paths
- Metrics (bandwidth, delay, etc.)
- Successors and feasible successors
But only the best routes make it into the routing table.
EIGRP Message Types
EIGRP uses several message types to communicate between routers:
- Hello: Used to discover and maintain neighbor relationships.
- Update: Used to send routing information.
- Acknowledgment (ACK): Confirms reliable receipt of an Update.
- Query: Sent when a route is lost and no feasible successor exists.
- Reply: Sent in response to a Query, containing route information.
What’s cool is that EIGRP is selective about how it communicates. Unlike RIP, which floods updates periodically, EIGRP only sends updates when something changes. That reduces bandwidth usage and improves efficiency.
EIGRP Timers
You don’t have to tweak timers often, but it’s good to know how they work.
- Hello Timer: How often EIGRP sends Hello packets.
- Hold Timer: How long a router waits before declaring a neighbor down.
- Retransmission Timeout (RTO): How long to wait before resending a packet if no ACK is received.
You can fine-tune these settings for performance, especially on WAN links.
Autonomous Systems in EIGRP
EIGRP is organized around autonomous systems (AS). When you configure EIGRP on a router, you assign it an AS number:
- router eigrp 100
All routers within the same AS will exchange routing information with each other. This is similar to how BGP uses AS numbers, but on a smaller, internal scale.
EIGRP for IPv6
EIGRP works with IPv6 too. The configuration is a bit different. Instead of using a global router process like you do in IPv4, you configure EIGRP for IPv6 on each interface:
- ipv6 router eigrp 1
- int g0/0
- ipv6 eigrp 1
Other differences include:
- No network statements (you enable it per interface)
- Uses link-local addresses for neighbor discovery
- Requires a router ID (manually assigned)
IPv6 support makes EIGRP a forward-looking protocol even though many enterprises are still slow to adopt IPv6.
EIGRP Route Types
Not all EIGRP routes are created equal. There are three main types:
- Internal Routes: Learned from other routers within the same AS.
- External Routes: Routes redistributed into EIGRP from another protocol like OSPF, RIP, or BGP.
- Summary Routes: Manually or automatically summarized routes.
External routes include a lot of extra info in the routing table, such as origin protocol and administrative distance.
Administrative Distance
EIGRP has an Administrative Distance (AD) of:
- 90 for internal routes
- 170 for external routes
That makes internal EIGRP routes more preferred than OSPF (which has an AD of 110), but external EIGRP routes are less preferred.
Why does this matter? Because when multiple protocols are in play, routers use the route with the lowest AD.
Load Balancing: Equal and Unequal
One of EIGRP’s coolest tricks is its support for unequal-cost load balancing. Most protocols only load balance across paths with the same cost. But EIGRP uses a command called variance to include less-than-perfect paths in the mix, as long as they meet certain criteria.
Here’s an example:
- router eigrp 1
- variance 2
This tells EIGRP to include paths with a metric up to twice as large as the best path, allowing for smarter traffic distribution.
EIGRP Configuration Basics (IPv4)
Let’s say you have three routers and want to enable EIGRP:
- router eigrp 100
- network 192.168.1.0
- network 192.168.2.0
That’s it! EIGRP will:
- Start sending Hello packets on those interfaces
- Discover neighbors
- Exchange routes
- Populate the topology and routing tables
Add no auto-summary if you’re dealing with discontiguous subnets, especially in modern CIDR-based networks.
Advantages of EIGRP
Let’s recap what makes EIGRP a solid choice:
- Rapid convergence thanks to DUAL
- Efficient use of bandwidth (no periodic updates)
- Loop-free routing
- Multiple path support, including unequal cost
- Scalable and flexible
- Supports both IPv4 and IPv6
It’s fast, it’s smart, and it’s easier to configure than OSPF in many cases.
Disadvantages of EIGRP
It’s not all sunshine and roses, though:
- Proprietary roots: While partially open, full compatibility across vendors is still limited.
- Less widely supported outside of Cisco ecosystems.
- Complexity: The metric calculation and DUAL algorithm can be harder to grasp compared to simpler protocols.
- Not the default in many environments: OSPF is more widely adopted in multi-vendor setups.
So, it’s perfect if you’re in a Cisco world. But maybe not the best fit in mixed environments.
Should You Use EIGRP?
If you’re working in a Cisco-only network (which is still very common in enterprises), EIGRP is a fantastic choice. It offers fast convergence, stable performance, and tons of features that simplify network management. If you’re studying for your Cisco certifications (CCNA, CCNP, or beyond) you’ll definitely want to get familiar with EIGRP. It shows up in labs, exams, and real-world scenarios all the time.
But if you’re in a mixed-vendor network, OSPF might be the safer bet since it’s an open standard and universally supported. Either way, understanding EIGRP gives you a deeper appreciation for how dynamic routing works behind the scenes. It’s not magic, it’s math, logic, and a little Cisco engineering finesse.
So whether you’re configuring it in your home lab or prepping for that big networking job, EIGRP is definitely worth knowing.
