Which best describes the fault tolerance of routing on the Internet?

• A. There are multiple routes for each IP packet to follow. A packet can be re-routed if a path becomes unavailable.
• B. Packets always follow the same precomputed path and cannot be rerouted.
• C. If a path fails, all packets are lost and never resequenced.
• D. Fault tolerance is unrelated to routing; it's a feature of end-host software.

Answer: (B)

Fault tolerance in routing comes from having many possible paths between endpoints and from routing protocols that keep updating which path to use. If a link or router along a route fails, routers can switch to alternate paths so packets still reach their destination. Because different packets can take different routes, a flow may arrive with packets out of order, which the transport layer or reassembly can handle. This is why the idea that every packet sticks to a single precomputed path forever or that a failed path causes all packets to be lost isn’t accurate. Routing is designed to adapt and keep delivery going even when parts of the network have issues.