[Note: Click the “DOWNLOAD” button to the right to download the free config files for this interactive lab]
Transcript: Welcome to this, our seventh CCDA lab. This time, we’ll be looking at some of the considerations for a scalable EIGRP design. We’ll consider things like the availability of physical successors and we’ll also look at the EIGRP stub feature.
Now in this lab you have three routers: R1, R2, and R3. R2 and R3 are connected to a switch so there’s a network there that we’re going to be using. That’s the 10.0.23 network slash 24, and then R1 and R2.
So let’s start with, let’s just go to console. So if we come here and we do a show IP route. So if we look at that network now, R1 knows about that network. That’s the 10.0.23 network. It knows about it. There are both R2 and R3, right? So if anything happens to any of these routers, traffic is still going to be forwarded and it will not have to query for this particular network, right? So if I also check my show IP EIGRP topology, so that’s the guide there.
So now for this lab I’m going to debug EIGRP. I like this FSM. So let’s look at the final state machine debugging. Let’s come here. What I’m going to do now is, I’m going to shut down this interface. That’s R1’s interface to R2. And then we’ll see what happens. So, that’s fa0/0 I believe and shut. Alright, so let this just load and then we’ll come back to it.
Okay, so it’s done. So Watch what happens now. It’s telling you that 2 went down and stuff like that. It’s saying that particular destination is entering the active state. Yes, so keep that in mind. So 10.12 is entering the active state. But notice that 10.0.23 has a network that [inaudible 00:02:06] via two parts did not enter active state. So because it was able to find a visible successor for that particular destination, can you see that, and then it just removed the other next hop and installed the one through R3. So that’s very important. Keep that in mind, it did not go into active state for 10.0.23.0. But it went into active state for this one because it did not have an alternate part for it already.
So let’s just, debug all, and I’m going to bring up that interface. Now the next error that I’m going to consider, in the first scenario, notice that these two guys have the same, see what’s like an equal cost. So look at this, it was an equal cost. But let’s do it in such a way that one would have a higher cost than the other. So if I go to fa0/1, that’s the 1 to error 3, I’m just going to play around with the delay. So let me just play around with the delay. If I check the topology table again, EIGRP topology, so right now it hasn’t changed. I’m expecting that to change. So yeah, that’s it. So this has increased, yeah, if you noticed that has increased. So if I do a show IP route only the part through 10.0.12.2 will be there. So let’s see that. So that’s that right there, cool. So right now it’s not equal cost. There’s no equal cost load balancing or anything like that. There’s no equal path but there is a physical successor. There is a feasible successor in the EIGRP topology table.
CCNA Quad Instant Pricing – Intense
Let’s do the same thing that we did. I’m going to debug again. Do debug EIGRP FSM. If I were to go to the fa0/0 interface and shut it down. That’s the interface to R2. So we’re going to shut it down. Now let’s wait for it to finish. Basically the same thing is going to happen. This particular destination is going to go to the active state. But, the destination 10.0.23.0 will not go into the active state because it was able to find a feasible successor for that particular destination. So all it’s going to do is to install it into the routing table.
So if I were to do a show IP route. Yeah, it knows that part’s through R3, right. So that’s the second scenario. In both cases there was a feasible successor. In the first one, there was equal path. In the second one, there was a feasible successor.
Now let’s see what happens if there’s no feasible successor. I’m going to debug all. I’m going to no shut this guy. I’m going to remove the delay on this guy. And I’m going to come here. And then its connection to that, so if I were to do show IP interface brief. Right so disconnection, I’m going to increase the delay on that guy. So interface fa0/0 let’s use a delay of say 2000. Now let’s see the effect on R1. If I were to check EIGRP. Now notice that it is not only released as a feasible successor, which is what I want to happen. Right now it is only saying the path through R2 so it doesn’t have a feasible successor for this particular destination now. So let’s see what happens again. I’m going to debug EIGRP of SM do debug. I’m going to go on that interface and shut it down.
Now notice what happens. Alright so it’s done. Let’s scroll back up. As always, it went into active state for the 10.0.12 but notice that it tried to find the feasible successor for 10.0.23.0 but it could not find because look at the reason it gave, and then it went into active state for that particular route. So that’s why when you design your networks, you should try as much as possible to have alternate paths to a particular network or something like that. It will reduce how much time your router is going to spend going into active states for a route that it doesn’t know. SO that’s the first thing that we are going to look at on that EIGRP.
The next thing we are going to look at is the stop feature. So I’m just going to stop this. In fact I’m just going to go to another project. So 7.2. We’re going to use this topology to talk about the EIGRP stop feature. Now it’s similar to the other topology that we had before, R1, R2, R3. But now, I have made some changes. So for example, if go show IP route, I can this 2.2 network here. Now this 2.2 network is from R2 and R3 knows about it through R2. But if I were to check the EIGRP topology, I would see that it also knows about it through its path through R1. So it knows about it through here, but it is not the best route so it’s using this one right now. But what would happen is, if this link goes down, R3 would try to reach R2 through R1 so let’s see that happen. See if I were to come here fa0/0, and say shut, now what would happen is it’s going to know about that route through R1. If I were to ping 2.2.2, it’s going to go through R1.
But now imagine R1 is a stub or a spoke that doesn’t have enough bandwidth on this link so it shouldn’t be used as a transit. So one thing we could do on R1 is to come here and configurate as a stub. So router, EIGRP 10, and then I’m going to say EIGRP stub. Now watch what would happen. If I come to R3, and I do a show IP route, that network is not going to be there, right. Because the stub basically says don’t use me as a transit, well one of the things it says is don’t use me as a transit path. So if I were to bring that interface back up, now the route will be back. But even though its back, do a show IP EIGRP topology, the path through R1 will not be listed as a feasible successor. So that’s one of the ways to use EIGRP stop feature.