Hello @Teck Sing
The behavior you're experiencing can be explained by differences in how RIPv2 and RIPng handle route withdrawals when interfaces or network advertisements change. In your IPv4 setup using RIPv2, when you either remove the 192.168.1.0 network from RIP or shut down the loopback interface (Lo0) on R1, R1 immediately starts advertising the 192.168.1.1/32 route with a metric of 16. This is RIP’s way of signaling to its neighbors that the route is unreachable, as metric 16 in RIP indicates an infinite distance. R2 then receives this update and propagates the unreachable route to its neighbors, ensuring that all routers in the network quickly discard the now-invalid route. This behavior is standard for RIPv2, ensuring that route invalidation is rapid and consistent across the network to prevent routing loops or stale routes.
In RIPng, the behavior is somewhat different due to the protocol’s approach to route advertisements. When you shut down the loopback interface on R1, RIPng behaves similarly to RIPv2 by sending out an advertisement for the unreachable IPv6 route with a metric of 16. This informs R2 that the route is no longer valid, and R2 will treat it as such and propagate the change accordingly. However, when you remove the `ipv6 rip RIP enable` command from the loopback interface, R1 simply stops advertising the route entirely, rather than sending an advertisement with a metric of 16. In this scenario, the last valid route advertisement remains in R2’s routing table until RIPng’s route timeout expires. During this period, R2 can still reach the 2001:DB8:0:1::1 address, even though R1 is no longer actively advertising it. Eventually, when the RIPng route timers expire, the route will disappear from R2’s routing table.
The difference in behavior between RIPv2 and RIPng can be attributed to their handling of route withdrawals. In RIPv2, removing a route from the RIP configuration or shutting down an interface both trigger a triggered update with metric 16, which ensures quick invalidation of the route in neighboring routers. RIPng, however, does not send an update with metric 16 when the routing process for an interface is disabled. Instead, it simply stops advertising the route, relying on the route expiration timers in the neighboring routers to eventually clear the stale route. This delay can lead to a brief period where the route is still reachable from neighboring routers, even though the originating router is no longer advertising it.
Best regards
.ı|ı.ı|ı. If This Helps, Please Rate .ı|ı.ı|ı.
