Reference Guide

Configure BFD for BGP
In a BGP core network, BFD provides rapid detection of communication failures in BGP fast-forwarding paths between internal
BGP (iBGP) and external BGP (eBGP) peers for faster network reconvergence. BFD for BGP is supported on 1GE, 10GE, 40GE,
port-channel, and VLAN interfaces. BFD for BGP does not support IPv6 and the BGP multihop feature.
Prerequisites
Before configuring BFD for BGP, you must first configure the following settings:
1. Configure BGP on the routers that you want to interconnect, as described in Border Gateway Protocol IPv4 (BGPv4).
2. Enable fast fall-over for BGP neighbors to reduce convergence time (the neighbor fall-over command), as described
in BGP Fast Fall-Over.
Establishing Sessions with BGP Neighbors
Before configuring BFD for BGP, you must first configure BGP on the routers that you want to interconnect.
For more information, refer to Border Gateway Protocol IPv4 (BGPv4).
For example, the following illustration shows a sample BFD configuration on Router 1 and Router 2 that use eBGP in a transit
network to interconnect AS1 and AS2. The eBGP routers exchange information with each other as well as with iBGP routers to
maintain connectivity and accessibility within each autonomous system.
Figure 14. Establishing Sessions with BGP Neighbors
The sample configuration shows alternative ways to establish a BFD session with a BGP neighbor:
● By establishing BFD sessions with all neighbors discovered by BGP (the bfd all-neighbors command).
● By establishing a BFD session with a specified BGP neighbor (the neighbor {ip-address | peer-group-name}
bfd command)
Bidirectional Forwarding Detection (BFD)
117