With OSPF providing full loopback reachability across the core, this video introduces BGP — starting with internal BGP between PE1 and PE2, establishing the first iBGP session in the topology and building the foundation for full-mesh iBGP across all four PE routers.

Before configuring BGP the video explains the design decision to use OSPF rather than ISIS as the IGP going forward. The OSPF configuration files from the previous series are available for download, and engineers coming from the ISIS video series can continue using ISIS as their IGP — the BGP configuration is identical regardless of which protocol is providing loopback reachability.

BGP configuration on JunOS starts with setting the autonomous system number under routing-options. For this topology all PE devices are in autonomous system 65000. BGP neighbours are grouped — all iBGP sessions go into a group named internal with the type set to internal. Because it is an iBGP group, the autonomous system number does not need to be specified per neighbour — Junos assumes all members of an internal group share the same AS.

The local-address is set to the device’s loopback address — 1.1.1.1 on PE1 and 2.2.2.2 on PE2 — because iBGP sessions should be sourced from the loopback rather than interface addresses. This means if any single link fails but OSPF can still reach the loopback via an alternative path, the iBGP session stays up. It’s the standard design for iBGP in any real network.

Once committed on both ends we verify with show bgp neighbor and confirm the session state is Established on port 179. The video explains the key fields in the BGP neighbour output — local address, peer address, autonomous system, session state, and the active/received/advertised prefix counts. At this stage no routes are being exchanged — iBGP is up but BGP only advertises what you explicitly tell it to advertise, unlike OSPF which shares everything in its area automatically. Route advertisement policy configuration comes in later videos.