Having discussed OSPF route summarization and the various types of stub areas, let’s look at the potential drawbacks of using those features. We’ll start by reviewing how OSPF acts before any summarization features are configured. Refer to Figure 1:
As you can see, R3 connects Area 0 and Area 1 together, making R3 an OSPF ABR (Area Border Router), and likewise for R4 (for fault-tolerance, it’s always a good idea to have two ABRs connecting a pair of areas). As part of their ABR function, both R3 and R4 will advertise each individual prefix within Area 0 into Area 1 (and vice-versa) using a Type 3 (Summary) LSA (Link State Advertisement). Each individual Type 3 LSA sent into Area 1 by the ABRs will appear in R5’s LSDB (Link State Data Base).
As we know, each OSPF router uses Dijkstra’s SPF (Shortest Path First) algorithm to determine the lowest-cost path to each prefix. For this simple topology, we can see by inspection that R5 (an OSPF internal router) will have two possible paths (via R3 or R4) to each prefix within Area 0.
Assuming that all of the links within the OSPF domain are Fast Ethernet, each link will have an OSPF cost of 1 (by default Cisco routers set the OSPF link cost equal to 100 Mbps divided by the bandwidth). Therefore, from R5’s perspective, we should have the following lowest-cost paths:
- Subnet A via R3 (cost = 3)
- Subnet B via R3 (cost = 3)
- Subnet C via R4 (cost = 3)
- Subnet D via R4 (cost = 3)
Note that all four subnets can also be reached the “long way around” (for example, Subnet A via R4 at a cost of 4), but those paths would not be used if a lower-cost path exists. The best path to each prefix within Area 0 will appear in R5’s IP routing table as an “O IA” (OSPF Inter-Area) route.
Next time, we’ll discuss how R5 sees the prefixes within the RIP and EIGRP clouds.
Author: Al Friebe