EIGRP Path Selection
EIGRP Path Selection: How DUAL Chooses the Best Route (and its Backups)
Enhanced Interior Gateway Routing Protocol (EIGRP) is renowned for its fast convergence, powered by the Diffusing Update Algorithm (DUAL). Understanding how EIGRP selects primary and backup paths is essential for learners, interview preparation, and network engineers aiming to optimize path selection and stability.
Part 1: The EIGRP Topology Table — The Brains of Path Selection
Unlike traditional distance-vector protocols, EIGRP maintains a Topology Table. This table contains all known destination networks within the EIGRP Autonomous System (AS) and tracks metrics received from neighbors.
| Term | Technical Definition (The “Why”) |
|---|---|
| Reported Distance (RD) | Metric advertised by a neighbor for a specific destination. Represents the neighbor’s calculated best metric to reach that destination. |
| Feasible Distance (FD) | The metric for the path with the lowest overall cost to a destination, calculated locally. Records the smallest known metric since the route last entered Passive state. |
| Successor Route | The route with the lowest path metric to reach a destination. It is the primary path for forwarding traffic. |
| Successor | The first next-hop router for the Successor Route. |
Goal: Calculate FD for every known network and identify the best next-hop (Successor).
Part 2: The Logic of Loop-Freedom — The Feasibility Condition (FC)
EIGRP ensures loop-free backup paths using the Feasibility Condition (FC).
Feasibility Condition Rule:
A neighboring router qualifies as a Feasible Successor (FS) if:
Reported Distance (RD) < Feasible Distance (FD)
Why RD < FD guarantees loop-freedom:
- The neighbor offering the backup path is a downstream router.
- Its metric is smaller than the local router’s historical best metric, so traffic moves closer to the destination.
- Routes meeting FC are guaranteed loop-free and eligible as backup paths.
- Neighbors not satisfying FC may be upstream and are not used as backups but are stored in the Topology Table.
Part 3: Path Selection and Convergence in Action
EIGRP uses DUAL states to maintain stability and enable fast convergence:
Passive State (Stable)
- Route is stable and in Passive (P) state if the Successor meets FC.
- If the primary path fails, the router checks for Feasible Successors.
- If a FS exists, the route remains Passive and switches immediately to FS — fastest convergence.
Active State (Recalculation)
- Occurs when the primary path fails and no Feasible Successor is available.
- The router enters Active (A) state, sending QUERY packets to neighbors to find a loop-free path.
- If a route stays Active beyond the Active Timer (default 3 minutes), it is declared Stuck in Active (SIA).
- Using FC minimizes Active state occurrences, improving network stability and speed.
Part 4: Beyond the Best Path — Load Balancing
EIGRP provides flexible load balancing options:
1. Equal-Cost Multipathing (ECMP)
- Multiple successor routes with identical metrics can be installed in the Routing Information Base (RIB).
- Default maximum paths: 4, adjustable via
maximum-pathscommand.
2. Unequal-Cost Load Balancing
- Feasible Successors with different metrics can be used safely.
- Achieved with the variance multiplier:
Install FS if FD <= Successor FD × variance
- FS paths are already loop-free, so unequal-cost load balancing is safe.
Key Takeaways:
- EIGRP uses DUAL to calculate Feasible Distance (FD) and choose Successors.
- Feasible Condition (FC) ensures loop-free backups.
- Passive/Active states control fast convergence.
- Supports ECMP and unequal-cost load balancing via variance.
- Topology Table stores all routes, including upstream routes, ensuring efficient path selection.
Pro Tip: Understanding RD, FD, Successors, FS, and FC is crucial for both network troubleshooting and ENARSI exam preparation.
🙌 Connect With Me
