OSPF is an interior gateway protocol (IGP) based on the the link-state concept of IP routing.
OSPF communicates by forming adjacencies with neighbour routers, using hello packets to form and maintain adjacencies, and link state advertisements (LSAs) to share routing information.
The nature in which link-state protocols operate specifies that each router within the OSPF domain maintains a link-state database (LSDB), which is shared between neighbour routers.
Changes to the network topology are communicated between neighours; this process is known within OSPF as synchronisation.
OSPF deals with scalability by using a hierarchical design. The design uses the concept of areas, with a single backbone area used to share summarised routing information between between all other non-backbone areas.
The relationship between different areas is an important aspect of OSPF design, and the many functions built into the protocol are fundamental to support the hierarchical nature of a multi-area OSPF domain.
This study of OSPFv2 will focus on the advanced levels of implementation that would be required for the JNCIE-SP certification.
Each topic listed below will be covered in the following sections, and at the end of the the four sections a lab will simulate each feature:
- Packet types and LSAs
- SPF and the Dijkstra algorithm
- Neighbour establishment and adjacencies
- Timers and timer optimisation
- Area types
- Router types
- Loop-free alternate routes (LFA)
- Load balancing
- MPLS extensions and CSPF support
The links and publications below are also excellent references that cover the entire protocol in detail, and should be referred to for further reading.
Juniper Tech Library (OSPF)
OSPF: Anatomy of Internet Routing Protocol
OSPF and ISIS: Choosing an IGP for Large Scale Networks