Configuring a BGP LSP

Context

Before a BGP LSP is established, a routing policy must be configured to control label distribution. The egress of the BGP LSP to be established needs to distribute an MPLS label to the route advertised to an upstream node. If a transit node on the BGP LSP receives a labeled IPv4 route from its downstream node, the transit node must re-distribute an MPLS label to the labeled IPv4 route before advertising the route to an upstream node.

Procedure

Perform the following steps on each CSG and MASG:
1. Run system-view
  The system view is displayed.
2. Run route-policy route-policy-name permit node node
  A Route-Policy node is created.
3. Run apply mpls-label
  The local device is enabled to distribute MPLS labels to IPv4 routes.
  
  By default, the action of distributing MPLS labels to public routes is not configured in the routing policy.
4. Run quit
  Return to the system view.
5. Run bgp { as-number-plain | as-number-dot }
  The BGP view is displayed.
6. Run peer { ipv4-address | group-name } route-policy route-policy-name export
  A routing policy for advertising routes matching Route-Policy conditions to a BGP peer (group) is configured.
  
  By default, no routing policy for filtering routes from its BGP peer (group) or routes to be advertised to its BGP peer (group) is configured.
  
  Repeat this step for each BGP peer.
7. Run network ip-address [ mask | mask-length ] [ route-policy route-policy-name ]
  The route destined for the loopback interface address is advertised.
  
  This route is labeled and advertised along a path over which a BGP LSP is established.
8. (Optional) Run ingress-lsp trigger route-policy route-policy-name
  The function to create ingress LSPs for labeled BGP routes based on routing policies is configured.
  
  By default, ingress LSPs are created based on all received BGP labeled routes.
  
  On a MAN where the hybrid access mode is used, a large number of labeled BGP routes are used to establish end-to-end LSPs. On certain intermediate nodes where VPN services do not need to be supported, excessive ingress LSPs are created, which waste network resources. In this case, you can run the ingress-lsp trigger command to create ingress LSPs based on a routing policy to save network resources.
9. (Optional) Run peer { group-name | ipv4-address } mpls-local-ifnet disable
  The ability to establish an MPLS local IFNET tunnel between a CSG and MASG is disabled.
  
  In the inter-AS seamless MPLS network transmitting L3VPN services, a CSG and MASG establish an MP-EBGP peer relationship. Therefore, an MPLS local IFNET tunnel between the CSG and MASG is established over the MP-EBGP peer relationship. The MPLS local IFNET tunnel fails to transmit traffic because the CSG and MASG are indirectly connected.
  
  If a fault occurs on the BGP LSP between the CSG and MASG, traffic is iterated to the MPLS local IFNET tunnel, not an FRR bypass tunnel. As the MPLS local IFNET tunnel cannot forward traffic, traffic is interrupted. To prevent the traffic interruption, run the peer mpls-local-ifnet disable command to disable the establishment of an MPLS local IFNET tunnel between the CSG and MASG.
Perform the following steps on each AGG, AGG ASBR, and core ASBR:
1. Run system-view
  The system view is displayed.
2. Run route-policy route-policy-name permit node node
  A Route-Policy node is created.
3. Run if-match mpls-label
  A matching rule based on the MPLS label is created.
  
  By default, no matching rule based on the MPLS label is configured.
4. Run apply mpls-label
  The local device is enabled to distribute MPLS labels to IPv4 routes.
  
  By default, the action of distributing MPLS labels to public routes is not configured in the routing policy.
5. Run quit
  Return to the system view.
6. Run bgp { as-number-plain | as-number-dot }
  The BGP view is displayed.
7. Run peer { ipv4-address | group-name } route-policy route-policy-name export
  A routing policy for advertising routes matching Route-Policy conditions to a BGP peer (group) is configured.
  
  By default, no routing policy for filtering routes from its BGP peer (group) or routes to be advertised to its BGP peer (group) is configured.
  
  Repeat this step for each BGP peer.