By default, in a typical configuration, BGP considers only IP routes for next-hop resolution when building a routing table. When an MPLS network uses BGP to propagate routes, BGP may consider whether the MPLS tunnels are viable routes.
You can globally enable BGP shortcuts using next-hop MPLS on an Extreme device to configure BGP to use an MPLS tunnel as the preferred next-hop route to a destination network when the tunnel is available. You can also configure BGP to include LSP metrics for best-route computations. You can also configure BGP to use the IGP metric of LSP tunnels for best-route computations. When the BGP attempt at route next-hop resolution through MPLS tunnels is unsuccessful, the Extreme device uses the IPv4 routing table to resolve the route.
When you configure BGP shortcuts using next-hop MPLS on an MPLS edge router, BGP computes routes to destinations available through other edge routers. When BGP determines that a route is available through an edge router that is reachable through an MPLS tunnel, a BGP shortcut directs BGP to place the MPLS tunnel in the routing table as the preferred BGP route.
Note
When an MPLS LSP tunnel to a BGP next hop is available, BGP always prefers the LSP tunnel to resolve a BGP next hop if the next-hop-mpls command is enabled. Mixed ECMP between MPLS LSP and native IGP path are not supported for a given BGP next hop.