EIGRP stub routing

NightShade1

Okay

It is normally done in a hub and spoke topology, its done to improve network stability

Okay the hubs are the distribution routers, and the spokes are the remote routers which are configured as Stubs ok i get that

The remote routers i mean the ones configured as stub can be conected to one or many hub routers(distribution routers)
Each distribution router can be connected to MANY spoke ones(the ones with stub networks)

The remote router(spoke) must forward all nonlocal traffic to a distribution router, so it becomes unnecessary for the remote router to hold a complete routing table Also it says

"The distribution router need not send anything more than a default route to the remote router."
What does this mean in here? i know that the remote router if he wanted to go outside of his local network he will just need a default route maybe? installed in his routing table, but what default route will the distribution router sent it? does it mean like a default network propagated by a routing protocol?


So whats the idea of using stub routing? so router wont learn routes from other networks? cause he just need to have in his routing table a default route for the distribution router? also to not recive any queries so bandwich and memory can be conserved <--- is this the idea of using EIGRP stub routing? i got it? or not?

now that will be in a simple hub and spoke with one spoke and with one hub

But now it seems different when many hubs and spokes are involved!

How the remote router can decide which is the best path to go for example in this one


how can the remote router decide which route is the best to go to the corporate network if it supposed that it doesnt have the full routing table to save memory and bandwich ? it supposed that it just got the default route It willl use the first default route listed in the routing table?

ill just ask this for now but i got more questions im just really confused witht he explanation given in cisco site.... i just undesrtand like a part of it and i dont for the other but i would like to undesrtand it complely so thats why im asking you guys en light me plz!

the main part of my misunderstanding is that i dont understand what routers got in their routing tables cause of the stub routing and how the sent packets i thinks thats what im missing

marlon23

Routers does not send DUAL queries to stub routers, but stub routers still send their routing tables, updates and queries. Which routes are send is based on stub configuration (receive-only, connected, static, summary; default is connected+summary).

So... Stub routers have all routing data as normal, but they are not asked with queries and you can set what data they will advertise.

Benefits (only examples..) :
- Reduce amount of traffic over WAN links = smaller updates send by stub routers + reduced amount of queries; more important is that you don't want hub routers to sen traffic between them through remote routers - just configure stub routers to only advertise for example connected networks, so it would not advertise for example default route and so)
- Improved stability of network = if for example WAN link to stub router goes down, hub router would not send queries to stub router, because these would not make it, and route would be in SIA state - Stuck In Active, it is better to use reliable core network for such a thing
- Improved convergence = hub routers wouldn't use slow WAN links for queries, they can use much faster core network to found missing routes



Hope it helps you to get some answers it is better to use some nice book than a cisco site when you want understand things easily