until the graceful restart completes. By exiting the graceful restart on a
topology change, a router tries to eliminate the loops or black holes as quickly
as possible by routing around the restarting router.
A helpful neighbor considers a link down with the restarting router to be a
topology change, regardless of the strict LSA checking configuration.
Use the nsf restart-interval command to configure the length of the grace
period on the restarting router. Use the "no" form of this command to revert
the grace period to its default.
nsf [ietf ] restart-interval
no nsf [ietf] restart-interval
ietf — This keyword is used to distinguish the IETF standard
implementation of graceful restart from other implementations. Since the
IETF implementation is the only one supported, this keyword is optional.
seconds — The number of seconds that the restarting router asks its
neighbors to wait before exiting helper mode. The restarting router
includes the restart interval in its grace LSAs (range 1–1800 seconds).
The default restart interval is 120 seconds.
The grace period must be set long enough to allow the restarting router to
reestablish all of its adjacencies and complete a full database exchange with
each of those neighbors.
console(config-router)#nsf restart-interval 180