thank you for your reply, from my understanding, router will share the reported AD from other router which means when R3 receives (AD=25 from R1) and (AD=19 from R2) so as a result --> R3 should send both AD’s to R4 and this does not make sense!
Hi Ziad,
In reality, when R3 receives an update from R1 or R2 about the destination behind R4, it won’t install them since these don’t pass the feasibility condition (AD of the feasible successor has to be lower than FD of successor).
R3 will only advertise its successor route to other neighbors. In this topology, R4 is the successor route so normally R3 would advertise this route to R4. However, because of split horizon (don’t advertise a route to a neighbor if you learned that route from the neighbor)…this route is not advertised.
It might be helpful to see all of this in action. Connect four routers like in the topology picture, then load these four configs:
R1-show-run-2018-01-17-15-58-39-clean.txt (270 Bytes)
R2-show-run-2018-01-17-15-58-43-clean.txt (270 Bytes)
R3-show-run-2018-01-17-15-58-46-clean.txt (366 Bytes)
R4-show-run-2018-01-17-15-58-50-clean.txt (261 Bytes)
If you want to see the EIGRP updates, you can start Wireshark and use this filter:
eigrp.opcode == 1
It will show you the route that is advertised and the metric.
Seeing this in action probably makes it easier to understand.
Rene
1 Like
Thank you for the clarification 
, i will try to demonstrate this test in lab , when i see the eigrp exchange packets in live i will get the full picture.
Hi Rene, The topic is very interesting, Thank you.
Hello Rene
can you tell me link where i find below topic
LISP encapsulation principles supporting EIGRP OTP
Hi Prashant,
It’s here:
Enjoy 
Rene
In the introduction of EIGRP it is mention that it is CISCO propitiatory. But now it is open standard. I think you have not updated it. I am confused now about the other content also.
hi rene would you create video for ad and fd for second network diagram
Hello Kapil,
You are correct, I originally wrote this lesson before Cisco made EIGRP an open standard. I just changed the text in the lesson. What other content are you confused about?
Rene
Hi Rudhra,
That might be a good idea, we get quite some questions about this diagram.
Rene
Hello Rene,
What is Eigrp hello and dead interval?
I am bit confused about it.
Is it different for LAN and WAN?
Thanks,
Swapnil k
Hello Swapnil
The hello timer or interval is the rate at which hello packets are sent from routers that are participating in EIGRP. On an Ethernet interface, the default hello timer is 5 seconds while on a serial interface, T1, ISDN, Frame relay and other such technologies, this value is 60.
Now EIGRP doesn’t have a dead interval. What I believe you mean is the hold timer. This is the amount of time that a router will consider a neighbor alive without receiving a hello packet. This is usually three times the hello interval. By default for Ethernet connections this value is 15 seconds while for other technologies as mentioned above, it is 180 seconds. If the hold time elapses without a hello packet being received, the neighbor is considered dead.
I hope this has been helpful!
Laz
With regards to the picture below. In the explanation for why R1 and R2 cannot be feasible successors because there would be a loop, I am confused as to why there would be a loop if either R1 or R2 were feasible successors. I understand that EIGRP has the condition that AD must be less than FD. What if R4 becomes unreachable and by default the destination network becomes unreachable. How would there be a loop if there is no network to go? In any case, if R4 is gone then the route to the destination would be removed from the routing table and no loop would take place. Is the explanation that there would be no feasible successor on R3 because there is no way to get to the destination if the destination is gone because R4 is down?
The topology is being viewed from the standpoint of R3. We have the benefit here of seeing the full network topology, and you can trace if R3 used R1 or R2 as a feasible successor, to reach the destination, you’ll be going in a loop back to R3 (where you started).
1 Like
I have a question, doing the eigrp lab for CCNA in gns3vault, in the next point Configure EIGRP AS 1 on both fastethernet links connecting router Bates and oHara , I find that for all the networks learned through the bates router I have two successors except for the loopback 1.1.1.0 belonging to the bates router. which shows me that he announces two identical ADs for both interfaces but the FD of the OHARA interfaces is different only for that network (they have the same BW and delay) (these are the only ones that are enabled).
Also at the time of changing the bandwidth in f1 / 0 so that it is only the successor route through F0 / 0, I change the successor for all routes except for the bats network 1.1.1.0/24, which is still the route through the f1 / 0 interface.
I attached the output of the show ip eigrp topoloy in ohara before changing the bw and after the change of the bw:
before
_**R2#sh ip eigrp topology**_
_**IP-EIGRP Topology Table for AS(1)/ID(192.168.21.2)**_
_**Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,**_
_** r - reply Status, s - sia Status**_
_**P 1.1.1.0/24, 1 successors, FD is 156160**_
_** via 192.168.21.1 (156160/128256), FastEthernet1/0**_
_** via 192.168.12.1 (386560/128256), FastEthernet0/0**_
_**P 3.3.3.0/24, 2 successors, FD is 2300416**_
_** via 192.168.12.1 (2300416/2297856), FastEthernet0/0**_
_** via 192.168.21.1 (2300416/2297856), FastEthernet1/0**_
_**P 4.4.4.0/24, 2 successors, FD is 2300416**_
_** via 192.168.12.1 (2300416/2297856), FastEthernet0/0**_
_** via 192.168.21.1 (2300416/2297856), FastEthernet1/0**_
_**P 192.168.12.0/24, 1 successors, FD is 258560**_
_** via Connected, FastEthernet0/0**_
_**P 192.168.21.0/24, 1 successors, FD is 28160**_
_** via Connected, FastEthernet1/0**_
_**P 192.168.134.0/24, 2 successors, FD is 2172416**_
_** via 192.168.12.1 (2172416/2169856), FastEthernet0/0**_
_** via 192.168.21.1 (2172416/2169856), FastEthernet1/0**_
After change the BW
_**R2#sh ip eigrp topology**_
_**IP-EIGRP Topology Table for AS(1)/ID(192.168.21.2)**_
_**Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,**_
_** r - reply Status, s - sia Status**_
_**P 1.1.1.0/24, 1 successors, FD is 156160**_
_** via 192.168.12.1 (386560/128256), FastEthernet0/0**_
_** via 192.168.21.1 (25730560/128256), FastEthernet1/0**_
_**P 3.3.3.0/24, 1 successors, FD is 2300416**_
_** via 192.168.12.1 (2300416/2297856), FastEthernet0/0**_
_** via 192.168.21.1 (26242560/2297856), FastEthernet1/0**_
_**P 4.4.4.0/24, 1 successors, FD is 2300416**_
_** via 192.168.12.1 (2300416/2297856), FastEthernet0/0**_
_** via 192.168.21.1 (26242560/2297856), FastEthernet1/0**_
_**P 192.168.12.0/24, 1 successors, FD is 258560**_
_** via Connected, FastEthernet0/0**_
_**P 192.168.21.0/24, 1 successors, FD is 25602560**_
_** via Connected, FastEthernet1/0**_
_** via 192.168.12.1 (261120/28160), FastEthernet0/0**_
_**P 192.168.134.0/24, 1 successors, FD is 2172416**_
_** via 192.168.12.1 (2172416/2169856), FastEthernet0/0**_
_** via 192.168.21.1 (26114560/2169856), FastEthernet1/0**_
_**R2#**_
I await your answers and recommendations thanks
In EIGRP by default there is a limited amount of Sucessor and FS by default for a path?, as in ospf and rip that there is a maximum of 4 paths by default.
regards
Hello Jesus
The Successor and Feasible Successor of EIGRP are different than the maximum paths used by OSPF and RIP. I’ll try to clarify below.
The successor is the route with the best metric to reach a particular destination. The feasible successor is the backup route that will immediately be promoted to the best route to a destination in the event that the successor fails. This is done without the need to rerun the DUAL algorithm.
Now the maximum paths used for OSPF refers to the maximum number of routes that can be installed in the routing table in the event that the routes have an equal cost. The default is 4 while the range can be configured between 1 and 16. EIGRP also has a similar maximum which is also set to 4 by default. Once again, this is for equal cost routes. So essentially, in such a case EIGRP can have up to four successors (by default) or best routes installed in the routing table. For EIGRP, this default value can be changed to anything between 1 and 6.
I hope this has been helpful!
Laz
One question, EIGRP only announces to the other neighbors its best route (sucessor)? in the explanation of the loop because R1 announces to R3 its route through R2 if that is not its best route, same thing happens with R2. I understand that they can not announce their best route by the split horizon rule
Hello Jesus
EIGRP does indeed only announce its best route to the neighboring routers. R1 announces its route to the destination to R2 and R2 in turn does the same to R3.
I hope this has been helpful!
Laz
I want to throw in a slight edit on your Introduction lesson here Rene. Specifically about EIGRP Split Horizon.
if the currently best route for a prefix list a particular outgoing interface. Split Horizon means that EIGRP will not include that prefix in the update sent out that same interface.
So basically i wont readvertise a route back out an interface on which that outgoing interface is the best path to reach that destination.
There are times when a downstream router you would think should receive multiple updates from its upstream peers about a destination but when you check the topology table it only knows of one route - the successor. What about the backup route you can clearly see on your visio diagram? Thats what i thought. The upstream neighbor will not advertise the link to its downstream neighbor if that link is the best path to reach that destination.
I would just slightly modify your split-horizon definition. IMO.