Cisco Express Forwarding
A key element of NSF is
packet forwarding. In Cisco networking devices, packet forwarding is provided
by Cisco Express Forwarding (CEF). Cisco Express Forwarding maintains the
Forwarding Information Base (FIB), and uses the FIB information that was
current at the time of the switchover to continue forwarding packets during a
switchover. This feature eliminates downtime during the switchover.
During the normal NSF
operation, Cisco Express Forwarding on the active fabric card synchronizes its
FIB and adjacency databases with the FIB and adjacency databases on the standby
fabric card. On switchover of the active fabric card, the standby fabric card
initially has FIB and adjacency databases that are mirror images of those on
the active fabric card. The packet forwarding continues after a switchover as
soon as the interfaces and a data path are available.
(NSF) works with the Stateful Switch Over (SSO) feature to minimize the amount
of time a network is unavailable following a switchover. The main objective of
the NSF is to continue forwarding IP packets after the switchover of the active
When a networking device
restarts, all the routing peers of that device usually detect that the device
went down and then came back up. This down-to-up transition results in routing
flap, which could spread across multiple routing domains. Routing flaps caused
by routing restarts create routing instabilities, which are detrimental to the
overall network performance. NSF helps to suppress routing flaps and improves
the network stability.
NSF allows for the
forwarding of data packets to continue along known routes while the routing
protocol information is being restored following a switchover. With NSF, peer
networking devices do not experience routing flaps. Data traffic is forwarded
through intelligent line cards while the standby fabric card assumes control
from the failed active fabric card during a switchover. The ability of line
cards to remain up during the switchover and to be kept current with the FIB on
the active fabric card is key to NSF operation. The CPT 50s connected to the
fabric card will have an impact on the traffic.
CPT does not support
forwarding IP packets in hardware and supports forwarding only in software.
The following figure
illustrates how SSO is typically deployed in service provider networks. In this
example, NSF with SSO is primarily at the access layer (edge) of the service
provider network. A fault at this point can result in loss of service for
enterprise customers requiring access to the service provider network.
Figure 1. NSF with SSO
Network Deployment: Service Provider Networks
Additional levels of
availability may be gained by deploying NSF with SSO at other points in the
network where a single point of failure exists. The following figure
illustrates an optional deployment strategy that applies NSF with SSO at the
enterprise network access layer. In this example, each access point in the
enterprise network represents another single point of failure in the network
design. In the event of a switchover or a planned software upgrade, enterprise
customer sessions would continue uninterrupted throughout the network.
Figure 2. NSF with SSO
Network Deployment: Enterprise Networks
The NSF feature has
the following benefits:
availability—NSF continues to forward network traffic and application state
information so that user session information is maintained after a switchover.
stability—Network stability may be improved with the reduction in the number of
route flaps that had been created when routers in the network failed and lost
their routing tables.
flaps—Because the NSF continues to forward network traffic in the event of a
switchover, routing flaps are avoided.
No loss of user
sessions—User sessions established prior to the switchover are maintained.
Each protocol depends
on Cisco Express Forwarding to continue forwarding packets during switchover
while the routing protocols rebuild the Routing Information Base (RIB) tables.
After the routing protocols have converged, Cisco Express Forwarding updates
the FIB table and removes stale route entries. Cisco Express Forwarding updates
the line cards with the new FIB information.
The routing protocols
run only on the active fabric card, and they receive routing updates from the
neighboring routers. After a switchover, the routing protocols request that the
NSF-aware neighboring devices send the state information to help rebuild the
During the NSF
operation, the routing protocols depend on Cisco Express Forwarding to continue
forwarding the packets while the routing protocols rebuild the routing
When an OSPF
NSF-capable router performs a fabric card switchover, it must perform two tasks
to resynchronize its link state database with its OSPF neighbors. Firstly, it
must relearn the available OSPF neighbors on the network without causing a
reset of the neighbor relationship. Secondly, it must reacquire the contents of
the link state database for the network.
After the fabric card
switchover, the NSF-capable router sends an OSPF NSF signal to the neighboring
NSF-aware devices. The neighboring network devices recognize this signal as a
clue that the neighbor relationship with this router must not be reset. As the
NSF-capable router receives signals from other routers on the network, it can
begin to rebuild its neighbor list.
relationships are reestablished, the NSF-capable router begins to resynchronize
its database with all of its NSF-aware neighbors. At this point, the routing
information is exchanged between the OSPF neighbors. After this exchange is
complete, the NSF-capable device uses the routing information to remove stale
routers and update the RIB and FIB with the new forwarding information. The
OSPF protocols are then fully converged.
The OSPF NSF
requires that all the neighboring network devices be NSF-aware. If a
NSF-capable router discovers that it has non NSF-aware neighbors on a
particular network segment, it will disable NSF capabilities for that segment.
Other network segments composed entirely of NSF-capable or NSF-aware routers
will continue to provide NSF capabilities.
Cisco had implemented
the proprietary Cisco NSF. The Graceful OSPF Restart feature supports IETF NSF
for OSPF processes in multivendor networks. The NSF device modes of operation
common to the Cisco and IETF NSF implementations are as follows:
this mode, the OSPF device is performing nonstop forwarding recovery because of
the fabric card switchover.
known as NSF–awareness. In this mode, the neighboring device is restarting and
helping in the NSF recovery.