Because the basic components of a VPLS network are the same regardless of whether BGP signaling or LDP signaling is used, the sample topology shown for LDP signaling in Figure 129 is almost identical to the sample topology shown for BGP signaling in Figure 128. Figure 129 includes two VPLS domains: VPLS A, which connects CE 1 and CE 3, and VPLS B, which connects CE 2 and CE 4. The local PE router, PE 1, and the remote PE router, PE 2, each participate in both the VPLS A domain and the VPLS B domain, and have one VPLS instance associated with each domain configured on each router.
Unlike a VPLS configuration with BGP signaling, a VPLS configuration with LDP signaling requires that you configure a VPLS ID for each VPLS instance to uniquely identify each VPLS domain. In the sample topology in Figure 129, instance vplsA is assigned VPLS ID 1, and instance vplsB is assigned VPLS ID 2 on both the local PE router and the remote PE router. You must also configure a list of remote neighbor (peer) addresses to which LDP can send or from which LDP can receive targeted hello messages. In the sample topology, the remote neighbor configured for PE 1 is PE 2 with IP address 18.104.22.168, and the remote neighbor configured for PE 2 is PE 1 with IP address 22.214.171.124.
The Ethernet and bridged Ethernet network interfaces, ATM core-facing interfaces, VPLS virtual core interfaces, and MPLS LSPs play the same role in a VPLS topology with LDP signaling as they do in a VPLS topology with BGP signaling. For more information about these components, see Topology Overview of VPLS with BGP Signaling.