Cisco Nonstop Forwarding (NSF) is a high availability initiative that enables routers to continuously forward IP packets following a route processor takeover or switchover to another route processor. NSF maintains and updates Layer 3 routing and forwarding information in the backup route processor. This ensures that the forwarding of IP packets and routing protocol information are continuous during the switchover and route convergence process. It eliminates router downtime, and increases network availability during scheduled maintenance of a route processor, or a route processor failure.

Cisco NSF, in conjunction with Cisco Stateful Switchover (SSO), has reduced the time it takes a router to respond to a failure of the main processor (also known as “failover”), from as much as half an hour to just 0-6 seconds. This new level of availability is making application reliability a guaranteed commodity end-to-end from service providers, rather than just to the edge. On many Cisco Internet-based router platforms, NSF and SSO together provide zero IP packet loss during an instantaneous route processor switchover, eliminating the aggregation router of a service provider as a single point of failure.

This is very significant, because a loss of connectivity at the aggregation edge-which often terminates hundreds of customers and thousands of sessions-has a direct impact on customer service level agreements (SLAs), often costing the service provider loss of business, loss of customer confidence, and possibly revenue in the form of rebates to customers. All routing peers of a device that restarts detect that the device went down and then came back up. In the past, the network might still be forwarding packets to the downed router, creating instabilities across multiple routing domains and impeding network performance. With Cisco NSF, all routers update their routing information tables when a route processor goes down. NSF allows the forwarding of data packets to continue along known routes while the routing protocol information is being restored following a switchover. Peer networking devices do not experience routing flaps. Traffic is forwarded through intelligent line cards or dual forwarding processors during a switchover.

Cisco NSF supports Border Gateway Protocol (BGP), Open Shortest Path First (OSPF), Intermediate System-to-Intermediate System (IS-IS), Enhanced Interior Gateway Routing Protocol (EIGRP) protocols for routing, and Cisco Express Forwarding for forwarding. Routers running these protocols can detect a switchover and take the necessary actions to continue forwarding network traffic and to recover route information from the peer devices.

Routing with NSF - The routing protocols run on the active route processor and receive routing updates from their neighbor routers. Following a switchover, the routing protocols request that the Cisco NSF-aware neighbor devices send state information to help rebuild the routing tables. Alternately, the IS-IS protocols can be configured to synchronize state information, from the active to the standby route processor, to help rebuild the routing table on the Cisco NSF-capable device in environments where neighbor devices are not Cisco NSF-aware. NSF extensions have been proposed to BGP, IS-IS, OSPF, and Enhanced Interior Gateway Routing Protocol (EIGRP).

Forwarding with NSF - A key element of Cisco NSF is Cisco Express Forwarding, which enables Cisco routers to forward packets. Cisco Express Forwarding maintains the forwarding information base (FIB) data that was current at the time of the switchover to continue forwarding packets to the new route processor.

For a more detailed explaination or for further reading regarding NSF and its interoperation with SSO, please see the following document: Cisco Nonstop Forwarding