Management and Control Overhead of Ibgp Network

Tutorial 6 (page 19 to continue)

Aims and Objectives

This tutorial will:

• Use Route Reflectors to simplify the configuration, management and control plane overhead of an iBGP network.

• Use Confederations to simplify the configuration, management and control plane overhead of an iBGP network.

• Use a route-map and a prefix list to modify the next hop addresses of targeted routes.

[pic 1]

• Build the topology shown on the previous page in VIRL.
• Download the initial configuration files for each router from SunSpace (or manually build the configuration files yourself!).
• Apply the configurations to the VIRL routers.
• The eleven configuration files contain the following:

• The IP addresses that need to be applied to each interface in the network.
• The names of each router
• The OSPF configuration that allows all routers in AS 2 to find the lo0 addresses of their peers.
• Static routes that enable eBGP peers to find the lo0 addresses of their peers.

Initial Tests

• Check that each router in AS 2 can reach the lo0 address of routers  R1 through to R9 from its own lo0 interface.
• Check that R6 and R7 can reach 11.11.11.11 from their own lo0 interface
• Check that R2 and R3 can reach 10.10.10.10 from their own lo0 interface.
• Check that each router in AS 2 can see all the connecting subnets (/30 subnets). You may use the command show ip route (see below)

[pic 2]

• Check that all iBGP interfaces are running OSPF and that each iBGP router has the expected number of OSPF neighbours. Ask if you are not sure how many neighbours and interfaces to expect for each router). The following screenshots contain the expected output from the commands show ip ospf neighbor and show ip ospf interface on R3.  [pic 3]

[pic 4][pic 5]

[pic 6][pic 7]

[pic 8]

BGP Configuration

If you were to configure iBGP in AS 2 as a full mesh of iBGP neighbours you would need to set up 36 separate neighbour relationships. The formula used to calculate this number is

N(N-1)/2

N is the number of iBGP peers which in this tutorial is 9. A full mesh of iBGP peers is not feasible as the network size grows. Fortunately, several alternative solutions have been devised. You will first of all look at the use of Route Reflectors to simplify the iBGP network.