Troubleshooting guide

Additional BGP Configuration

the external AS, packets destined for those networks will be dropped. A remedy for this scenario is for the

iBGP peer to advertise its own IPv4 address as the next-hop address to the external networks. Consider the

following example:

192.168.1.2

Router C

Router B

AS 65000

AS 65001

Router A

192.168.1.1

10.10.10.1

Figure 2. Using the Next-Hop-Self Command between iBGP Peers

Router B in AS 65001 has an eBGP neighbor relationship with Router A in AS 65000. All networks in AS

65000 are advertised from Router A to Router B with a next-hop IPv4 address of 10.10.10.1. Subsequently,

when Router B announces these networks to its iBGP neighbor (Router C), the BGP default setting is to

announce that the next hop to reach these networks is the entrance to AS 65000 (10.10.10.1). Router C

MUST have either a static route or an entry in its IGP routing table, indicating a route to reach the edge

router at 10.10.10.1. Otherwise, any information destined for networks in AS 65000 will be dropped by

Router C. The next-hop-self command can be issued on Router B so that Router B’s IP address

(192.168.1.1) is advertised to Router C as the next-hop address for networks in AS 65000. Since Router B

and C are directly connected, Router C’s routing table contains a route to Router B.

No Default Originate

The no default-originate co

mmand prevents the unit from sending the default route to a BGP neighbor.

This is the default setting for AOS devices.

(config-bgp-neighbor)#no default-originate

The transmission of default routes to BGP neighbors is accomplished by manually entering a default route

in the BGP Neighbor Configuration. This is done by entering the following command:

(config-bgp-neighbor)#network 0.0.0.0 mask 0.0.0.0

Example 4 on page 41 further demonstrates the use of the next-hop-self command.

This command cannot be enabled. In other words, the command default-originate cannot

be entered to enable the default route to be sent to a BGP neighbor.