This command is enabled by default on CISCO IOS. The command terminates external BGP sessions of any directly adjacent peer if the link used to reach the peer goes down; without waiting for the hold-down timer to expire.
Although this feature improves the BGP conversion time, it may lead to great instability in your BGP table due to a flapping interface.
The following example shows R1 & R2 with an EBGP session between them configured as follows:
|R1(config-if)#router bgp 100
R1(config-router)#nei 22.214.171.124 remot 200
R2(config)#router bgp 200
bgp fast-external-fallover enabled:
By shutting down the interface connecting between R1 & R2; R1 immediately resets the peering session without waiting for the hold-down time to expire as shown in the debug output below
*Mar 8 13:58:59.510: BGP: 126.96.36.199 resetting - interface Serial1/0 down
*Mar 8 13:58:59.578: BGPNSF state: 188.8.131.52 went from nsf_not_active to nsf_not_active
*Mar 8 13:58:59.582: BGP: 184.108.40.206 went from Established to Idle
*Mar 8 13:58:59.582: %BGP-5-ADJCHANGE: neighbor 220.127.116.11 Down Interface flap
*Mar 8 13:58:59.590: BGP: 18.104.22.168 closing
bgp fast-external-fallover disabled:
After shutting the interface connecting between R1 & R2 down the session remains up until the hold-down timer expires as shown in the debug output below
|R1(config)#router bgp 100
R1(config-router)#no bgp fast-external-fallover
One Final Note:
The command neighbor x.x.x.x fall-over can be used for fast deactivation of both EBGP or IBGP neighbors. This command is configured per neighbor and it terminates the session once the route to the neighbor is lost.