Information About OSPFv3 Nonstop Routing
OSPFv3 Nonstop Routing feature allows a device with redundant Route Processors (RPs) to maintain its Open Shortest Path First (OSPF) state and adjacencies across planned and unplanned RP switchovers. This feature works by checkpointing the OSPFv3 information from the active RP to the standby RP. When a changeover occurs and the standby RP becomes the new active RP, this checkpointed information is used to continue operation without interruption.
Although OSPFv3 Nonstop Routing serves a similar function to the OSPFv3 graceful restart feature, it works differently. With graceful restart, OSPFv3 on the newly active standby RP initially has no state information, so it uses extensions to the OSPFv3 protocol to recover its state from neighboring OSPFv3 devices. For this to work, the neighbors must support the graceful restart protocol extensions and be able to act as helpers to the restarting device. They must also continue forwarding data traffic to the restarting device while this recovery is taking place.
With nonstop routing, by contrast, the device performing the changeover preserves its state internally, and in most cases the neighbors are unaware that changeover has happened. Because no assistance is needed from neighboring devices, nonstop routing can be used in situations where graceful restart cannot; for example, graceful restart is unreliable in networks where not all the neighbors implement the graceful restart protocol extensions or where the network topology changes during recovery.
When nonstop routing is enabled, the responsiveness and scalability of OSPF is degraded. The performance degradation happens because OSPF uses CPU and memory to checkpoint data to the standby RP.