[Day 7] 6. Routing Loop

A routing loop happens when routing tables contain circular paths, causing packets to continuously circulate between routers without reaching their intended destination.

Example Scenario:

Network Setup:

• A → C → D → E
• B is connected to both A and C, forming a triangular topology.

Routing Loop Scenario:

1. Initial Routing:
   • Router A sends packets to C to reach E.
   • Router B might incorrectly route packets through A to reach E, creating a loop.
   • Router C forwards packets to D to reach E.
2. Loop Formation:
   • Due to a routing error or delay, A might keep sending packets to C.
   • C sends the packets to D.
   • If D has incorrect routing information, it might route packets back to C (possibly due to incorrect information from B), causing the packets to circulate between C and D.
   • Note: Delays in routing updates can result in outdated route information, contributing to the loop.

Solutions to Prevent Routing Loops:

1. Hop-Count Limit:

   • Description: Sets a maximum number of hops (e.g., 15 in RIP) to reach a destination. Any route exceeding this hop limit is considered unreachable, preventing the loop from continuing indefinitely.

2. Split Horizon:

   • Description: Prevents a router from sending routing updates back to the router from which it received the update. This helps eliminate loops caused by incorrect route advertisements.

3. Hold-Down Timers:

   • Description: Temporarily suppresses route changes for a specified period (typically 180 seconds) to stabilize the routing table and prevent constant route changes. During this period, the routing table remains stable, reducing the chance of routing loops due to fluctuating information.
   • Note: Split Horizon rules still apply during the hold-down period.

4. Triggered Updates:

   • Description: Sends updates immediately when a topology change occurs, rather than waiting for the regular update interval. This helps propagate changes quickly and reduces the risk of loops by ensuring routers have up-to-date information.

Additional Techniques for Loop Prevention:

1. Route Poisoning:

   • Description: When a router detects that a route is no longer valid, it marks that route as "invalid" by setting its hop count to a very high number (e.g., 16 in RIP), signaling to other routers that the route should no longer be used.

2. Poison Reverse:

   • Description: When a router learns that a route is invalid, it marks the route as invalid and informs the router from which it received the route. This helps prevent the invalid route from being used in a loop.

These techniques work together to improve the stability of routing protocols, preventing loops and ensuring that routing information remains accurate and up-to-date.