Questions: EIGRP: Enhanced Interior Gateway Routing Protocol

Question 1

Router A currently reaches destination X via neighbor B with a feasible distance (FD) of 100. Neighbor B reports an advertised distance (AD) of 70. Neighbor C reports an AD of 110. Which neighbor can serve as a feasible successor (loop-free backup)?

• A. Only C — it has a higher AD, meaning it has a longer path that could not be routing back through A
• B. Only B — its AD (70) is strictly less than the current FD (100), satisfying the feasibility condition and guaranteeing a loop-free alternate path
• C. Both B and C — any alternate path qualifies as a feasible successor
• D. Neither — a feasible successor can only be selected after B's route fails and DUAL queries are sent

Answer: (B)

DUAL's feasibility condition requires that the neighbor's advertised distance (AD) be strictly less than the current feasible distance (FD). If neighbor B's AD < FD, it is provably not routing through A to reach X — it has a genuinely shorter independent path. Neighbor C's AD (110) exceeds the FD (100), meaning C might be routing through A, which could create a loop. The feasibility condition is EIGRP's guarantee of loop-free routing at every instant: B can be used as an instant failover; C cannot without re-querying.

Question 2

A network engineer needs to minimize routing protocol bandwidth overhead on a WAN link. Why would she prefer EIGRP over RIP?

• A. EIGRP uses a simpler hop-count metric that requires less computation to advertise
• B. EIGRP sends only incremental, triggered updates when topology changes occur — not periodic full routing-table broadcasts like RIP's 30-second updates
• C. EIGRP compresses its routing table before transmission, reducing packet size
• D. EIGRP automatically reduces update frequency when bandwidth is congested

Answer: (B)

RIP broadcasts its entire routing table to all neighbors every 30 seconds, regardless of whether anything changed. On a slow WAN link, this constant churn wastes precious bandwidth. EIGRP establishes neighbor relationships with lightweight hello packets (sent every 5 seconds by default) and only sends updates for specific routes that changed — triggered by actual topology changes. An EIGRP network with a stable topology may go hours or days with virtually no update traffic. This is one of EIGRP's most important operational advantages over RIP in real deployments.

Question 3

EIGRP's feasibility condition guarantees that a feasible successor cannot be routing traffic back through the querying router, making loop-free instant failover possible.

• T. True
• F. False

Answer: True

This is the mathematical insight behind DUAL. If neighbor B's advertised distance (the metric B reports from itself to destination X) is less than router A's current feasible distance to X, then B cannot possibly be routing through A to reach X — if it were, B's distance would have to include A's distance, making it at least as large as A's FD. The strict inequality (AD < FD) is the invariant that proves no loop exists. When the primary route fails, the feasible successor can be used immediately without any query-response cycle, enabling sub-second convergence.

Question 4

In EIGRP, the feasible distance (FD) for a route and the advertised distance (AD) reported by the successor router refer to the same metric value.

• T. True
• F. False

Answer: False

FD and AD are distinct values. The advertised distance (AD) is what the neighbor router reports — the metric from that neighbor to the destination. The feasible distance (FD) is the full metric from the local router to the destination, which equals the metric of the link to the neighbor plus the neighbor's AD. FD ≥ AD always (since FD includes additional path cost). This distinction is critical to the feasibility condition: a backup route is safe if its AD is less than the current FD, not less than the current AD.

Question 5

Explain how EIGRP's feasibility condition prevents routing loops during a topology change, and contrast this with how RIP handles the same situation.

Model answer: EIGRP pre-qualifies backup routes (feasible successors) before any failure occurs, using the feasibility condition (neighbor's AD < current FD). When a primary route fails, the router instantly switches to the feasible successor without querying other routers — no temporary loop can form because the backup was mathematically proven loop-free in advance. RIP has no such pre-qualification: when a route fails, neighboring routers may still advertise the old route to each other, causing count-to-infinity loops where routers increment the metric through each other indefinitely. RIP's slow-convergence fixes (split horizon, route poisoning, holddown timers) reduce but do not eliminate this risk, and they introduce delays. EIGRP trades computational complexity (maintaining FD/AD state) for guaranteed loop-freedom and fast convergence.

This is the fundamental difference between basic distance-vector routing and DUAL. RIP's simplicity comes at the cost of loop vulnerability during convergence. EIGRP's DUAL algorithm maintains enough state to guarantee loop-freedom at every moment — not just after convergence stabilizes. The 'stuck in active' problem is EIGRP's remaining weakness: when no feasible successor exists, the router must query neighbors, and in large flat networks this query diffusion can be slow.