EIGRP Stubs – Purpose and Benefits

eigrpospfroutingstubsubnet

I have a strong background in OSPF and am now learning EIGRP so everything I read gets compared to my existing knowledge of OSPF. Because of this, I don't see the purpose of EIGRP stubs.

In OSPF, you make an area stub so that it does not receive type 5 or 4 LSAs thus reducing the size of the LSDB and saving memory resources. You can make it a stub no-summary so that it doesn't receive type 3, 4 or 5 LSAs further reducing memory resources. This is a logical configuration because stub areas have only one entry point to the backbone network therefore it makes no sense for a stub candidate to learn hundreds of routes all to the same next hop of the ABR when you could simply make it a stub so that it only learns a default route and its intra-area routes.

EIGRP is different from this. In EIGRP, stubs aren't about limiting what routes you learn but rather what routes you advertise. I don't see the correlation between having a single entry point to the rest of the network (AKA a stub router) and limiting outbound route advertisements. Stub routers still need to advertise their routes out to the rest of the network – how does having a single entry point into the backbone EIGRP network make them eligible to not advertise their routes?

What benefit does EIGRP stub routing provide if any?

Best Answer

There are philosophical differences in the design of OSPF and EIGRP and they are reflected in the protocol operation. For example, OSPF relies on a strict hierarchy/star topology to prevent loops. EIGRP uses the concept of successors for loop prevention, and this allows for more flexible topologies (not always a good thing).

EDIT:

OSPF and EIGRP stubs solve somewhat different problems.

The purpose of OSPF areas is to save resources by limiting the flooding of route updates to just inside the area.

OSPF stubs save resources in the stub area by blocking external routes inside the area (except, of course, NSSA).

EIGRP stubs are confined to individual routers, not areas. They save resources by limiting the size of the DUAL route query, and therefore speeds up convergence. An EIGRP stub advertises its own routes, but doesn't advertise routes it learns from other parts of the network. An EIGRP stub router tells its neighbor that it doesn't know about the rest of the network, so there's no point querying it about other routes. That speeds up convergence by limiting the number of queries a router has to process.

Related Topic