03-30-2011 02:15 PM - edited 03-04-2019 11:56 AM
Hi Everybody,
I am studying EIGRP for my CCNP and there is some details that are not clear about EIGRP updates.
1) Anycast or Multicast
I have found a lots of documentation stating that update are unicast during neighbor discovery and multicast when there is a ne link active.
The problem is that this is not what I observed when I sniffed link between two routers using EIGRP routing.
Unicast update during neighbor discovery seems to be correct but update mutlicast for a new active link seems to be correct only when there is more than two router connected together (multi-acces segment).
Unicast update seems to be used for point to point link. What is the rules for EIGRP updates?
2) Updates are sent only to affected routers (Bounded)
Can somebody gives me more detail about affected routers. What are the affected routers, is it only for neighboring routers? If a neighboring routers cannot find any successor for a route, doest it have to generate some new updates to it`s own neighbors.
Thanks for your help
Stephane
Solved! Go to Solution.
03-30-2011 07:03 PM
Stephane
Let me suggest an example explaining bounded updates and affected routers. Let us assume 5 routers. R1 has loopback interface with address 192.168.1.1 and FastEther 0/1 with address 10.0.12.1/24. R1 runs EIGRP on all interfaces. R2 has loopback with 192.168.2.1/24, FastEther0/0 with 10.0.12.2, and FastEther0/1 with 10.0.23.2/24. R2 runs EIGRP on all interfaces. R3 has loopback with 192.168.3.1/24, FastEther0/0 with 10.0.23.3/24 and FastEther0/1 with 172.16.34.3/24. R3 runs EIGRP on all interfaces and summarizes network 10.0.0.0/8 out FastEther0/1. R4 has loopback 192.168.4.1/24 and FastEther0/0 with 172.168.34.4, and FastEther0/1 with 172.16.45.4/24. R4 runs EIGRP on all interfaces. R5 has FastEther0/0 with 172.16.45.5/24 and runs EIGRP on it.
Now assume that the link between R1 and R2 goes down. R2 sends an update to R3 removing 10.0.12.0/24. R3 is affected since it had an entry in it's routing table for 10.0.12.0. R3 removes the affected entry. R3 sends an update to R4. But R4 does not have an entry for 10.0.12.0, so R4 is not affected and R4 does not send an update to R5.
HTH
Rick
Sent from Cisco Technical Support iPhone App
03-30-2011 07:03 PM
Stephane
Let me suggest an example explaining bounded updates and affected routers. Let us assume 5 routers. R1 has loopback interface with address 192.168.1.1 and FastEther 0/1 with address 10.0.12.1/24. R1 runs EIGRP on all interfaces. R2 has loopback with 192.168.2.1/24, FastEther0/0 with 10.0.12.2, and FastEther0/1 with 10.0.23.2/24. R2 runs EIGRP on all interfaces. R3 has loopback with 192.168.3.1/24, FastEther0/0 with 10.0.23.3/24 and FastEther0/1 with 172.16.34.3/24. R3 runs EIGRP on all interfaces and summarizes network 10.0.0.0/8 out FastEther0/1. R4 has loopback 192.168.4.1/24 and FastEther0/0 with 172.168.34.4, and FastEther0/1 with 172.16.45.4/24. R4 runs EIGRP on all interfaces. R5 has FastEther0/0 with 172.16.45.5/24 and runs EIGRP on it.
Now assume that the link between R1 and R2 goes down. R2 sends an update to R3 removing 10.0.12.0/24. R3 is affected since it had an entry in it's routing table for 10.0.12.0. R3 removes the affected entry. R3 sends an update to R4. But R4 does not have an entry for 10.0.12.0, so R4 is not affected and R4 does not send an update to R5.
HTH
Rick
Sent from Cisco Technical Support iPhone App
04-04-2011 11:37 AM
Hi Rick,
Great example, very helpful....
Thanks
Stéphane
04-05-2011 02:54 PM
Stephane
I am glad that you found my explanation helpful. Thanks for marking the question as resolved (and thanks for the points). It makes the forum more useful when good responses are rated, since it helps readers to identify questions where they expect to find helpful responses. You have contributed to that process. so thank you.
HTH
Rick
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide