Configure and Verify eBGP (IPv4 and IPv6 address families)

Exam: Cisco 300-101 - CCNP Implementing Cisco IP Routing (ROUTE v2.0)


The Border Gateway Protocol (BGP) is a multi protocol that is capable of carrying the information about routed protocols such as IPv4 Multicast, IPv4 Unicast, IPv6 Multicast, and IPv6 Unicast called address families, in a single session.

Configuration of eBGP using IPv4 address family:

Ipv4 address family is used to identify the routing session for BGP protocol that uses standard IPv4 prefixes. Multicast and Unicast prefixes are specified within the IP address family. When the BGP peer is configured, the routing information for IPv4 unicast is advertised by default until the advertisement for its information is explicitly turned off.

The configuration steps for eBGP IPv4 are:

Enable the privilege mode

RouterR1 > enable

Enter the global configuration mode

RouterR1# config t

Configure the router with autonomous system (AS) using command router bgp. Number 65001 is used as autonomous system number.

RouterR1(config) #router bgp 65001

Configure the router with IPv4 address family and enter the address family configuration mode.

RouterR1 (config–router) # address-family ipv4

Exit the address family configuration mode and returns to privileged EXEC mode

RouterR1 (config–router-AF) # end

Configuration of eBGP using IPv6 address family:

The use of the IPv6 address family is to identify routing sessions for BGP protocol that uses IPv6 address prefixes. The configuration of ebgp for IPv6 does not require any new version. IPv6 is a standard version of BGP.The code example below shows how to take advantage of the IPv6 multiprotocol support.

The steps required to configure BGP using IPv6 are:

Enable the privilege mode

RouterR1 > enable

Enter the global configuration mode

RouterR1# config t

Configure the router with AS using command router bgp. Number 65004 is used as autonomous system number.

RouterR1(config) #router bgp 65004

Configure the router with IPv6 address of a neighboring router having an IPv6 address 2102:CA8:0:DD00:1:1 and the remote autonomous system (AS) number. 65005 represents the remote AS number.

RouterR1 (config–router) # neighbor 2102:CA8:0:DD00:1:1 remote-as 65005

Configure the router with IPv6 address family and enter the address family configuration mode.

RouterR1 (config–router) # address-family ipv6

Configure the router by enabling the neighbor using activate keyword to exchange prefixes for address family with a local router. Address 2102:CA8:0:DD00:1:1 represents the IPv6 address of the neighboring router.

RouterR1 (config–router-AF) # neighbor 2102:CA8:0:DD00:1:1 activate

Exit the address family configuration mode and return to privileged EXEC mode

RouterR1 (config–router-AF) # end

4 byte AS number

Earlier, BGP autonomous system (AS) number allocated to companies were 2 byte numbers ranging from 1 to 65535 but, due to increase in demand for AS numbers, the Internet Assigned Number Authority has started to allocate 4 bytes AS number ranging from 65536 to 4294967295. Cisco implements two methods to represent AS numbers. These methods are:

  • As plain – This is a decimal value notation where both two byte and four bytes AS numbers are represented by their decimal value.
  • As dot – This is an autonomous system dot notation where two byte AS numbers are represented by their decimal value and four bytes AS numbers represented by dot notation.

The four byte AS numbers can be configured in both AS plain format and dot format.

Use the following steps to configure the eBGP with 4 bytes AS number:

Enable the privilege mode

RouterR1 > enable

Enter the global configuration mode

RouterR1# config t

Defines the 4 byte autonomous system number using router bgp command. Number 65540 is represented as 4 byte AS number.

RouterR1 (config) #router bgp 65540

Configure the router with an IP address of neighbor, using the neighbor command having IP 192.168.3.8 and remote autonomous system number as 65539

RouterR1 (config–router) # neighbor 192.168.3.8 remote-as 65539

Configure the router with an IPv4 address family with unicast keyword which specifies the IPv4 unicast address family.

RouterR1 (config–router) # address family ipv4 unicast

Configure the router to enable the neighbor having an IP 192.168.6.4 to exchange prefixes for IPv4 unicast address family.

RouterR1 (config–router-AF) # neighbor 192.168.6.4 activate

Configures the router to specify the network having IP 10.10.10.1 and mask 255.255.255.0 as local autonomous system and add it to BGP routing table.

RouterR1 (config–router-AF) # network 10.10.10.1 mask 255.255.255.0

Exit the address family configuration mode and returns to privileged EXEC mode

RouterR1 (config–router-AF) # end

Private AS number

Private AS numbers are used by ISPs to preserve globally unique autonomous system (AS) numbers. Autonomous system numbers appear in eBGP AS paths in routing updates. Private AS numbers range from 64512 to 65535. As private AS numbers are not unique, they cannot leak to the global BGP routing table and to calculate the best path in BGP they require a unique autonomous system number. Therefore, it is necessary to remove private autonomous system numbers from AS path before routes are advertised to BGP peer.

When a customer network is configured with private AS number, the BGP updates from the customer network and the ISP will have private AS number in its AS path list. When an ISP advertise the information about network to the internet, it should not advertise the AS path with a private AS number of the customer to the internet.

Following configuration steps shows how to configure and remove private AS from AS path.

Enable the privilege mode

RouterR1 > enable

Enter the global configuration mode

RouterR1# config t

Configure the router as an eBGP neighbor having IP 192.168.3.1 in private AS. Number 65001 represents private AS number.

RouterR1 (config–router) # neighbor 192.168.3.1 remote as 65001

Removethe private AS number using command remove private as all from AS path in the outgoing eBGP update. IP 192.168.8.3 used as eBGP neighbor.

RouterR1 (config–router) #neighbor 192.168.8.3 remove-private-as all

Exit the current configuration mode and returns to privileged EXEC mode

RouterR1(config–router)# end

Verification Of eBGP

After configuring eBGP, there may be problems that you can face. For example the local and remote BGP peer may not able to reach each other or the IP address is not configured properly on a router. For that you need to verify the configuration. Use the following commands for verification.

Displays the entries in the routing table of BGP

Show ip bgp

Displays information about TCP and BGP connections with neighbors

Show ip bgp ipv4

Displays entries in IPv6 BGP routing table

Show bgp ipv6 unicast

Displays status of all IPv6 BGP connections

Show bgp ipv6 unicast summary

Displays the status of all BGP connections

Show ip bgp summary

Displays the information about a BGP update group

Show ip bgp update-group

Displays the information about BGP neighbors

Show ip bgp neighbors

Exam Question:

Which statement is true about eBGP?

  1. An internal routing protocol can be used to reach an EBGP neighbor.
  2. A static route can be used to form an adjacency between neighbors.
  3. The next hop does not change when BGP updates are exchanged between EBGP neighbors.
  4. EBGP requires a full mesh.

Answer: 2

Explanation: When two eBGP neighbor wants to establish a neighboring relationship without using the directly connected interfaces, they must tell each other how to reach their interfaces. For that, a static route is the most simple way.


Related IT Guides

  1. Configure and verify device management
  2. Configure and Verify OSPF for IPv6
  3. Configure and verify policy-based routing
  4. Configure and verify switch administration
  5. Configure and verify tracking objects
  6. Explain Frame Relay
  7. Explain TCP operations
  8. Identify Cisco Express Forwarding concepts
  9. Identify IPv6 addressing and subnetting
  10. Identify, configure, and verify IPv4 addressing and subnetting