Deployment Guide
Conguring BGP Fast Fail-Over
By default, a BGP session is governed by the hold time.
BGP routers typically carry large routing tables, so frequent session resets are not desirable. The BGP fast fail-over feature reduces the
convergence time while maintaining stability. The connection to a BGP peer is immediately reset if a link to a directly connected external
peer fails.
When you enable fail-over, BGP tracks IP reachability to the peer remote address and the peer local address. Whenever either address
becomes unreachable (for example, no active route exists in the routing table for peer IPv6 destinations/local address), BGP brings down
the session with the peer.
The BGP fast fail-over feature is congured on a per-neighbor or peer-group basis and is disabled by default.
To enable the BGP fast fail-over feature, use the following command.
To disable fast fail-over, use the [no] neighbor [neighbor | peer-group] fail-over command in CONFIGURATION
ROUTER BGP mode.
• Enable BGP Fast Fail-Over.
CONFIG-ROUTER-BGP mode
neighbor {ip-address | peer-group-name} fail-over
Example of Verifying that Fast Fail-Over is Enabled on a BGP Neighbor
Example of Verifying that Fast Fail-Over is Enabled on a Peer-Group
To verify fast fail-over is enabled on a particular BGP neighbor, use the show ip bgp neighbors command. Because fast fail-over is
disabled by default, it appears only if it has been enabled (shown in bold).
Dell#sh ip bgp neighbors
BGP neighbor is 100.100.100.100, remote AS 65517, internal link
Member of peer-group test for session parameters
BGP version 4, remote router ID 30.30.30.5
BGP state ESTABLISHED, in this state for 00:19:15
Last read 00:00:15, last write 00:00:06
Hold time is 180, keepalive interval is 60 seconds
Received 52 messages, 0 notifications, 0 in queue
Sent 45 messages, 5 notifications, 0 in queue
Received 6 updates, Sent 0 updates
Route refresh request: received 0, sent 0
Minimum time between advertisement runs is 5 seconds
Minimum time before advertisements start is 0 seconds
Capabilities received from neighbor for IPv4 Unicast :
MULTIPROTO_EXT(1)
ROUTE_REFRESH(2)
CISCO_ROUTE_REFRESH(128)
Capabilities advertised to neighbor for IPv4 Unicast :
MULTIPROTO_EXT(1)
ROUTE_REFRESH(2)
CISCO_ROUTE_REFRESH(128)
fail-over enabled
Update source set to Loopback 0
Peer active in peer-group outbound optimization
For address family: IPv4 Unicast
BGP table version 52, neighbor version 52
4 accepted prefixes consume 16 bytes
Prefix advertised 0, denied 0, withdrawn 0
192
Border Gateway Protocol IPv4 (BGPv4)