What a static route actually is
A static route is an administrator manually entering a destination network and how to reach it, either a next-hop IP (Internet Protocol) address or an outgoing interface. That's the entire mechanism: no protocol runs in the background, no negotiation with neighboring routers, nothing gets learned automatically.
Where it's the right tool
- Small networks with few routers and paths that basically never change.
- A default route pointing a stub network's only exit toward its single upstream connection.
- Point-to-point links where there's only ever one possible path anyway, so a dynamic protocol has nothing to compute.
Where it falls apart
Static routes don't adapt. If the configured next-hop goes down, the route stays in the table (unless the router also loses reachability to the next-hop itself) and traffic keeps getting sent into a dead end until a human notices and fixes it. And every network has to be entered by hand, on every router that needs to reach it. A topology with dozens of routers becomes an enormous, error-prone amount of manual configuration. That's exactly the problem the dynamic routing protocols later in this module exist to solve.
Working example: A floating static route as a backup ISP link
The floating static route from the previous topic, worked through concretely. R1 has two paths out to the internet: a primary link through ISP-A, and a backup link through ISP-B that should only ever carry traffic if the primary is actually down.
Two static default routes point at the same destination, 0.0.0.0/0, but with different administrative distances:
ip route 0.0.0.0/0 203.0.113.1: the primary route, left at its default administrative distance of 1.ip route 0.0.0.0/0 198.51.100.1 5: the backup route, with its administrative distance manually raised to 5.
As long as R1 can reach 203.0.113.1, the AD-1 route wins outright (see Administrative Distance & Metrics), and the AD-5 route never gets installed in the routing table at all, even though it's sitting there fully configured. The moment the primary link actually goes down, R1 can no longer reach 203.0.113.1, so that route disappears from the table on its own. With nothing left to outrank it, the AD-5 route becomes the best match and gets installed automatically: no script, no human, no dynamic protocol involved, just two static routes and the same administrative distance comparison from the previous topic.
A floating static route only fails over when R1 actually loses its route to the next-hop, typically because the outgoing interface itself goes down. If the physical link stays up but the ISP beyond it silently stops forwarding traffic, R1 has no way to know: the AD-1 route stays installed and "working" as far as the routing table is concerned, and the backup never kicks in. Real deployments usually pair this with a reachability check like IP SLA tracking to catch that case too.