What Is RIP?

The Original Distance-Vector Protocol

RIP — the Routing Information Protocol — is an interior routing protocol, and a distance-vector one. That term captures its whole philosophy: a router thinks about each destination as a distance (how far away) in a vector (which direction, i.e. which neighbor to send through). The distance RIP uses is the simplest metric imaginable — the hop count, the number of routers a packet must cross to reach the destination.

Critically, a RIP router does not see the whole network. It knows only what its direct neighbors tell it, and it periodically shares its entire routing table with those neighbors. Each router takes what it hears, adds one hop, and updates its own table. Routes propagate across the network like gossip passed person to person — which is both the charm and the downfall of the design.

How RIP Learns Routes

The mechanism is refreshingly simple, and you can follow it in a sentence each:

1. Every router periodically broadcasts its full routing table to its directly connected neighbors.
2. A neighbor receives the table and, for each destination, adds 1 to the advertised hop count (it is one hop further away from the neighbor's perspective).
3. If that path is better (fewer hops) than what the router already has — or is a new destination — it updates its table and points the route toward the neighbor that advertised it.
4. The router then advertises its updated table onward, and the information spreads outward one hop per update cycle.

There is no map, no calculation of link quality — just "my neighbor can reach X in 3 hops, so I can reach X in 4 hops through that neighbor." A router picks whichever neighbor offers the lowest hop count to each destination. Simple to implement, easy to understand, and exactly why it struggles at scale.

The 15-Hop Limit and Why It Exists

RIP's most famous quirk is that it considers a hop count of 16 to mean infinity — unreachable. In practice this caps the network at 15 hops; anything farther simply cannot be reached. This is not an arbitrary small number, it is a deliberate safety valve.

The problem it guards against is the counting-to-infinity failure. When a link goes down, the gossip nature of RIP can let routers feed each other stale information: router A loses a route, but hears from router B that B can still reach it (because B has not yet learned the link failed), so A re-learns the now-broken route through B at a higher hop count. B then hears it back from A, increments again, and the count spirals upward — a loop where the hop count climbs endlessly while packets circle. Capping at 16 stops the spiral: once a route reaches 16, every router agrees it is dead and discards it. The cost of this fix is the hard 15-hop ceiling, which confines RIP to small networks.

Slow Convergence — the Real Weakness

Beyond the size limit, RIP's deeper flaw is slow convergence. Convergence is how quickly a network settles on correct routes after something changes. Because RIP routers share updates only periodically and learn of changes secondhand, hop by hop, news of a failed link seeps across the network gradually rather than propagating instantly. During that window, different routers hold contradictory views, and temporary loops can form until everything stabilizes.

RIP added several patches to soften this — techniques with names like split horizon (don't advertise a route back to the neighbor you learned it from), route poisoning (actively advertise a dead route as unreachable), and hold-down timers (ignore updates about a route for a while after it changes). These help, but they treat symptoms of the fundamental design: a router acting on gossip rather than a real map can never converge as fast as one that sees the whole topology.

Why OSPF Replaced It

RIP's limitations map almost one-to-one onto the strengths of OSPF, the link-state protocol that displaced it. Where RIP knows only hop counts from neighbors, OSPF gives every router the full network topology and lets it compute the genuinely best path by cost — so a fast multi-hop route beats a slow single hop, which RIP could never recognise. Where RIP converges slowly and caps at 15 hops, OSPF converges quickly and scales to large networks through areas.

Today RIP survives mainly in very small or legacy networks and as a teaching tool — its simplicity makes it the perfect introduction to dynamic routing before tackling more capable protocols. It earned its place in history; it just no longer earns a place in most new designs.

Frequently Asked Questions

What is RIP?

The Routing Information Protocol, one of the oldest interior routing protocols. It is distance-vector: each router measures distance by hop count and periodically shares its whole routing table with neighbors, building routes from what neighbors report rather than from a full view of the network.

Why does RIP have a 15-hop limit?

RIP treats 16 hops as infinity — unreachable — to stop routing loops from counting upward forever. When a route's hop count hits 16, routers discard it. The side effect is that anything more than 15 hops away is unreachable, limiting RIP to small networks.

What is the difference between RIP and OSPF?

RIP is distance-vector using hop count, sharing its table with neighbors and converging slowly. OSPF is link-state — every router learns the whole topology and computes paths by cost, converging quickly and scaling far larger. OSPF makes better path choices and largely replaced RIP.

Is RIP still used?

It is largely obsolete in serious networks, replaced by OSPF and others. It survives mainly in very small or simple networks, in legacy equipment, and as a teaching tool, since its simplicity makes distance-vector routing easy to learn. New deployments rarely choose it.

What does slow convergence mean in RIP?

Convergence is how long a network takes to agree on correct routes after a change. RIP converges slowly because routers share updates only periodically and learn changes secondhand, hop by hop, so a failure spreads gradually. During the delay routers can hold stale routes, risking temporary loops.