EIGRP Neighbors and Adjacencies

mzinz

I have always thought that for two routers to become "neighbors", they had to be directly connected.

After reading some, it now seems that any two routers in the same AS have the opportunity to become neighbors.

Is that true?

On top of that--EIGRP routers share their topology tables with their neighbor adjacencies, right? But don't they just send it out on a multicast address? Wouldn't that cause all EIGRP routers in the AS to pick it up, even if the adjacency hasn't been established?

Any help would be great, I'm beginning to confuse myself.

NightShade1

You are like lost dude...
No they must be directly connected so they can be neighbors
They dont share routing tables for example EIGRP neighbors exchange all the routes known by the sending router which mean they exchange topology table information they DONT exchange routing table information.... the routing table is made of the topology table, in the topology table you got all the routes and metrics and it picks the best one and put it on the routing table and the other routes stay as back up in the topology table, if the route in the routing table fails then the back up route will take place in the routing table.....
On OSPF happens something similar... it has all the routes in the topology table and then it calculate the best path with th SPF algorithm and then it place it on the routing table.

Look Vector distance protocols such as RIP v1 and IGRP does share their routing tables, they depend of the neighbor for the information... They sent their routing update by broadcasting their entire routing table to all neighbors.

IGRP is better than rip even if they both vector distance protocol just because the period of update is longer which means it will be not soo CPU intensive like RIP also the update packet is a way more efficient

EIGRP are also a vector distance protocol but it does not broadcast and does not sent all the routing table and are not periodic either.

Link state protocol like OSPF has his view of the whole network itself i mean every router has the view of the whole network(topology table)
They work like this
1)they establish a neighboring with a hello packet and then they establish adjacency after this so neighbor discovering is the first step,

2)After that they start a synchronization of their database(topology table) to check they have the same database
3)After the synchronization of the database they start something called the LSA flooding. Every LSA(link state advertisement) is copied and re sent it to every neighbor except for the one that sent it...
And what are LSA? well they are like the updates that the topology table needs to fill up ....There are like 8 types or LSAs if i remenber well.

OSPF designate something called DR(designated router) and BDR(back up of the designated router)
In every broadcast multiaccess network such as ethernet there will be a BR and BDR election
What is the funtion of the DR ? well the function is the sent all the LSAs to all the routers in the same broadcast network.... the idea with this is to reduce the routing traffic because that router will act like a central point so all router must have full adjacenty to those DR and BDR
There is away lot of stuff you can do with OSPF but i wont type it...

EIGRP works different than OSPF it also has the neighbor, topology and routing table but works different....

they establish adjacency with neighbor and neighbor sent all the routes known by him

I don't think i can actually explain this one to you and i don't even think i should ve tried to explain you OSPF but well i already typed it so i will just post it...
Anyways go and read a book if something that you reading you don't understand ill explain you
Now what i don't understand is how you passed CCNA thinking that OSPF or EIGRP share their routing tables.... they share the status of links in their updates, vector distance protocols(RIP) complete routing table information.