Configuring inter-AS option C
To configure inter-AS option C, configure PEs and ASBRs.
PE configuration:
Configure basic MPLS L3VPN, and specify the PE in another AS as an MP-EBGP peer. The route targets for the VPN instances on the PEs in different ASs must match for the same VPN.
Execute the peer ebgp-max-hop command to enable the local router to establish an EBGP session to an indirectly-connected peer because the PEs are not directly connected.
Specify the ASBR in the same AS as an IBGP peer, and enable BGP to exchange labeled IPv4 unicast routes with the ASBR.
ASBR configuration:
Configure a routing protocol, and enable MPLS and LDP on the interface connecting to an internal router of the AS.
Specify the PE in the same AS as an IBGP peer, and the ASBR in a different AS as an EBGP peer.
Enable BGP to exchange labeled IPv4 unicast routes with the PE in the same AS and the ASBR in different AS.
Enable MPLS on the interface connected to the remote ASBR. There is no need to configure a label distribution protocol, for example, MPLS LDP.
Configure a routing policy to determine which IPv4 unicast routes are advertised to the IBGP or EBGP peer with MPLS labels.
In addition, configure BGP to advertise routes destined for a PE on PEs or ASBRs. For more information, see Layer 3—IP Routing Configuration Guide.
Configuring a PE
For basic MPLS L3VPN configurations on a PE, see "Configuring basic MPLS L3VPN." The following table describes inter-AS option C specific configurations.
To configure a PE for inter-AS option C:
Step | Command | Remarks |
|---|---|---|
1. Enter system view. | system-view | N/A |
2. Enter BGP view. | bgp as-number | N/A |
3. Configure the ASBR in the same AS as an IBGP peer. | peer { group-name | ip-address } as-number as-number | By default, no BGP peer is created. |
4. Configure the PE of another AS as an EBGP peer. | peer { group-name | ip-address } as-number as-number | By default, no BGP peer is created. |
5. Create the BGP IPv4 unicast address family and enter its view. | address-family ipv4 [ unicast ] | By default, the BGP IPv4 unicast address family is not created. |
6. Enable BGP to exchange IPv4 unicast routes with the ASBR in the same AS. | peer { group-name | ip-address } enable | By default, BGP does not exchange IPv4 unicast routes with any peer. |
7. Enable BGP to exchange labeled IPv4 routes with the ASBR in the same AS. | peer { group-name | ip-address } label-route-capability | By default, BGP cannot exchange labeled routes with any IPv4 peer or peer group. |
8. Return to BGP view. | quit | N/A |
9. Enter BGP VPNv4 address family view. | address-family vpnv4 | N/A |
10. Enable BGP to exchange VPNv4 routes with the PE in different ASs. | peer { group-name | ip-address } enable | By default, BGP cannot exchange VPNv4 routes with any peer. |
11. (Optional.) Configure the PE to not change the next hop of routes advertised to the EBGP peer. | peer { group-name | ip-address } next-hop-invariable | Configure this command on the RR so the RR does not change the next hop of advertised VPNv4 routes. |
Configuring an ASBR-PE
To set up an inter-AS public tunnel for the inter-AS option C solution, an ASBR must assign an MPLS label to the route destined for a PE, and advertise the label along with the route. Typically, the routes advertised by an ASBR through BGP include the PE address as well as other routes. You can configure a routing policy to filter routes. Routes surviving the filtering are assigned labels, and all others are advertised as common IPv4 routes.
To configure a routing policy, use the following commands:
if-match mpls-label—Matches routes carrying MPLS labels.
apply mpls-label—Sets MPLS labels for IPv4 routes to be advertised to a peer. You can use this command together with if-match clauses. For example, when used together with the if-match mpls-label command, the apply mpls-label sets new MPLS labels for routes with MPLS labels. The newly assigned labels are advertised along with the routes.
For more information about routing policy configuration, see Layer 3—IP Routing Configuration Guide.
To configure an ASBR-PE for inter-AS option C:
Step | Command | Remarks |
|---|---|---|
1. Enter system view. | system-view | N/A |
2. (Optional.) Create a routing policy, and enter routing policy view. | route-policy route-policy-name { deny | permit } node node-number | By default, no routing policy is created. |
3. (Optional.) Match IPv4 routes carrying labels. | if-match mpls-label | By default, no MPLS label match criterion is configured. |
4. (Optional.) Set labels for IPv4 routes. | apply mpls-label | By default, no MPLS label is set for IPv4 routes. |
5. Return to system view. | quit | N/A |
6. Enter BGP view. | bgp as-number | N/A |
7. Configure the PE in the same AS as an IBGP peer. | peer { group-name | ip-address } as-number as-number | By default, no BGP peer is created. |
8. Configure the ASBR in another AS as an EBGP peer. | peer { group-name | ip-address } as-number as-number | By default, no BGP peer is created. |
9. Create the BGP IPv4 unicast address family and enter its view. | address-family ipv4 [ unicast ] | By default, the BGP IPv4 unicast address family is not created. |
10. Enable exchange of IPv4 unicast routes with the PE in the same AS and the ASBR in another AS. | peer { group-name | ip-address } enable | By default, BGP does not exchange IPv4 unicast routes with any peer. |
11. Enable exchange of labeled IPv4 routes with the PE in the same AS and the ASBR in another AS. | peer { group-name | ip-address } label-route-capability | By default, BGP cannot advertise labeled routes to any IPv4 peer or peer group. |
12. Configure the ASBR-PE to set itself as the next hop of routes advertised to the PE in the local AS. | peer { group-name | ip-address } next-hop-local | By default, BGP does not use its address as the next hop of routes advertised to an IBGP peer or peer group. |
13. (Optional.) Apply a routing policy to routes incoming from or outgoing to a peer or peer group. | peer { group-name | ip-address } route-policy route-policy-name { export | import } | By default, no routing policy is applied. |
Configuring a routing policy on an ASBR-PE
A routing policy on an ASBR-PE does the following:
Assigns MPLS labels to routes received from the PEs in the local AS before advertising them to the peer ASBR-PE.
Assigns new MPLS labels to labeled IPv4 routes advertised to PEs in the local AS.
Which IPv4 routes are assigned with MPLS labels depends on the routing policy. Only routes that meet the criteria are assigned with labels. All other routes are still common IPv4 routes.
To configure a routing policy for inter-AS option C on an ASBR-PE:
Step | Command | Remarks |
|---|---|---|
1. Enter system view. | system-view | N/A |
2. Create a routing policy and enter routing policy view. | route-policy route-policy-name { deny | permit } node node-number | By default, no routing policy is created. |
3. Match IPv4 routes carrying labels. | if-match mpls-label | By default, no match criterion is configured. |
4. Set labels for IPv4 routes. | apply mpls-label | By default, no apply clause is configured. |