Hi. Welcome to this CertificationKits CCNA Training video on Configuring the Frame Relay. We are going to be talking about point-to-point versus multipoint interfaces, inverse ARP, what that does, static IP to DLCI mapping, sample configurations, we are going to go over a few different sample configurations, the configuration commands we will see, as well as some troubleshooting.
First thing I want to talk about here is the point-to-point versus multipoint interfaces. Now with frame relay at the CCNA level, we have the opportunity to have one physical interface, one line, connect to multiple other devices. So router 1 could connect directly to router 2 and router 3 simultaneously through this one line here. That would be considered a multipoint interface.
This router, if we look at this scenario, we might have router A, router B, and there is only one router connected to one other router, that would be considered a point-to-point interface. And it’s important to understand the differences here mainly because of IP addressing as well as the way it’s going to handle broadcast packets. But with the addressing in an environment like this where we have multipoint interfaces on each of the routers, it’s a full mesh environment we are looking at right here.
Every interface on each router that’s part of this frame relay set up would share one subnet. If we had to set up over here with router A and maybe another router, router C, so router A had a connection over to router C as well as router B, but there was no connection between router B and router C, we would have two point-to-point subinterfaces. We will take a look at what the subinterface does for us in a moment when we look at configuring. But one connection between A and B is one virtual circuit. The other connection between A and C would be another virtual circuit. So this would be one, two different subnets, and what we will do is we will create subinterfaces one for each subnet. We will go in and take a look at what I mean by creating those subinterfaces in a moment.
For now, let’s go in and take a look at what inverse ARP will do for us. Inverse ARP allows us to map our DLCI’s next top IP addresses. It’s a dynamic mapping. Just like if I had that phone in my office, this is a phone I am drawing here in case you are wondering, so here is the phone. I have got three lines on the phone, line 1, line 2 and line 3, and I need to know who is at the other end of those particular lines. That way, if I want to talk to, let’s say, Bob, I know which line to pick up.
So how that would work is I would pick up the line and send an inverse ARP out and out of my phone and that would end up on Bob’s phone over here. And if he heard me say, “Hi, this is Jason,” and the information came in on line 1, he would go, “Okay, so I want to talk to Jason, I will pick up line 1.” And he would send his inverse ARP back to me and say, “This is Bob,” and that information might come in on line 2 on mine. And so I go, “Okay, if I want to talk to Bob, I am going to pick up line 2.” Same thing with the inverse ARPS.
What happens is our router sends us inverse ARP message out automatically, and this default configuration sends an inverse ARP out, what will happen is DLCI 10 is a local DLCI over here for router B, what will happen is basically I am saying, “Hey, I am saying router 172.16.1.1. How are you?” And it will go, “Okay, 172.16.1.1,” that’s all in DLCI 10. And if we had multiple other routers, that’s where it can come in real handy so we knew what virtual circuit would get us to what router.
So he might have another virtual circuit, let’s say, VC 30 gets them over here to router C. This is router A, and there might be an IP address here of 172.16.2.1. And this guy sends his inverse ARP, and he goes, “Oh, okay, I have another IP address 172.16.2.1, and that’s DLCI of 30. So if I have a packet destination or destination subnet 172.16.10.0 and I need to send it to the next top router of 172.16.2.1, I am going to send it out of virtual circuit with a DLCI of 30.”
You wouldn’t know that information without inverse ARP. We can statically map it if we want to, and we are going to take a look at doing that. But again, inverse ARP allows us to map an IP address to a DLCI, and again B would send his inverse ARP out to router A, and you go, “Oh, 172.16.1.2, I want to send information to that next top. I will use DLCI, and that’s 20.”
So that’s what inverse ARP will do for us, and then take a look at the sample config – so one other thing we need to go over and you need to know, Frame Relay Forum, FRF, has a couple of standards. One standard is FRF 5; the other standard is FRF 8. FRF 5 is when in the WAN cloud, you have got ATM and then frame relay protocol local loop on both ends, so frame relay here, frame relay over here. FRF 8 is when one side is frame relay local loop between DTE and the DCE, and all of the rest of it is ATM, so just a couple of different standards FRF 5, FRF 8, I would know those two terms. Most of the time, I personally don’t care what’s going on in the WAN cloud; I am worried about local loop, unless you are working for a service provider, then you probably do care what’s going in the WAN cloud.
So just a couple of terms you get to need to know for the CCNA. What I want to do now is go in and take a look at some sample configurations for different topologies. We are going to be looking at point to point, P to P; we are going to be looking at a full mesh environment and a partial mesh. So let’s go in and take a look at those CCNA level configurations.
All right, I have brought up a full mesh environment. Full mesh means A has a connection to B; B has a connection to C. I am going to go with the global DLCI, so the global map of the DLCI, 10 represents A; 20 represents router B, and 30 represents router C. In a situation where you have a full mesh environment, everybody connects to everybody. We can actually configure frame relay on a physical interface. If we did not have a full mesh, everybody connecting to everybody, we would have to create subinterfaces like Sr0.1 would be one subinterface. And we will take a look at configuring those when we do our point-to-point in partial mesh.
I type encapsulation to frame relay. If I hit enter, it uses that Cisco type field. If you remember from the previous video with frame relay, all the frame relay goodies getting to know frame relay, if we have our IP packet LAPF, Link Access Procedure Frame bearer services, that’s a fun one to remember, specifies the header and the trailer. Now there is that little type field right here that tells the receiving device what layer 3 protocol was used. That type field has two options, Cisco or IETF, Internet Engineering Task Force. This is the multi-vendor one, taking guess which one is proprietary CISCO, this one. So if we just type encapsulation frame relay, hit enter, it will use the CISCO type field. And I just give it an IP address 172.16.1.1.
Same thing for Palaestra B, Sr0, encapsulation frame relay, enter, give it an IP of 1.2. And then for Palaestra C, encapsulation frame relay, IP of 1.3. Notice, all of the IP addresses are in the 172.16.1.0 subnet. So they all share one subnet. So it can be really easy to go in and configure frame relay. If we have a full mesh, we configure everything with defaults. Some of those defaults would be LMI type autosensed. So whatever our switch is using here, remember the LMI status messages between us and the DCE, the frame relay switch, that’s a service provider, that’s autosensed. Remember, there are three types of LMI, Cisco being the Cisco proprietary one if you have got all Cisco equipment, and then you’ve got ANSI and q933a. Those are different LMI types autosensed.
Permanent virtual circuit DLCI information is learned via the status messages. Again, what’s going on with what virtual circuits we have like router B has two virtual circuits, one with a DLCI of 10 so this one is DLCI of 10 because it gets them to Palaestra A which has a global DLCI of 10 and one with a DLCI of 30 which gets them to Palaestra C with a global DLCI of30. Now how he finds out what next top address is associated with these DLCIs is that inverse ARP. A sends out an inverse ARP in both directions. Router B gets it and goes, “Oh, if I go out of virtual circuit with a DLCI of 10, I will get to next top router 1.1. If I go out of virtual circuit with a DLCI of 30, I will get to next top router 1.3, and they will all use those inverse ARPs to get that information.”
So again, it can be pretty easy because LMI types autosensed what virtual circuits and DLCIs you have all autosensed with the status messages from the LMIs inverse ARP allows us to map our next top IPs with the DLCIs, and the type field is Cisco which is a proprietary one by typing in encapsulation frame relay enter, it uses a proprietary Cisco type field. So that’s a pretty easy one. Let’s take a look real quick how we would manually map the DLCI to the IP address if we wanted to turn off inverse ARPs. So let’s take a look at that, and then we will go in and we will take a look at a partial mesh point-to-point setup.
All right, I have brought up full mesh, same one we just looked at, but the difference is I turned off inverse ARP, and we are just looking at Palaestra A’s configuration. So DLCI global of 10, 20 and 30, IP addresses of 1.1, 1.2, 1.3, 172.16.1.0 subnet. They are all sharing it. And so what you do is you use the command no frame relay inverse arp, and that turns off inverse ARP. Since inverse ARP is off, we have to map DLCI to the IP address, and this is the command, frame relay map ip 172.16.1.2 is the next top address which would be router B and DLCI of 20. So this DLCI is going to be 20 right here representing this virtual circuit. And if router A sends something out of the virtual circuit with a DLCI of 20, it will end up at the IP address 1.2 for the next top.
Broadcast allows the router to simulate the broadcast messages. So any message that’s a broadcast that would normally go out of the Sr0 interface would be turned into unicast messages and sent out to each virtual circuit. So gain, I would know this command frame relay map ip, next top address DLCI and then broadcast allows the simulation of the broadcast message. This is next top; this is the protocol, pretty much going to be IP most of time and then frame relay map. This says frame relay; this says map. So I would definitely know the components of this command.
As a CCNA, let’s go in and take a look at a point-to-point environment and how we would configure that. Let’s take a look at an environment with a point-to-point setup. So here is Sr0 on the Palaestra Central router. And as a global DLCI of 10, we have Palaestra A, Palaestra B and Palaestra C global DLCI of 20, 30 an d40 respectively. We are going to look at Palaestra Central’s configuration first. So since we are going to need to create subinterfaces on Sr0, we get rid of the IP address; we go no IP address encapsulation frame relay. And remember, we don’t see it, but it’s specified in Cisco as the type field there.
Then to create the subinterface, we are basically chopping Sr0 up into multiple virtual interfaces, virtual interface 0.1, 0.2 and 0.3, and it actually acts like three separate physical interfaces. So to create the subinterface, that’s it, we just type it in interface serial 0.1 and then we have to specify the type of interface it's going to be, point-to-point or multipoint. In this situation, central connects to A, central connects to B, central connects to C but there is no connection between A and B or B and C or anything like that. So everything routes through the central router. Once we have the subinterface created, we specify the IP address, and in this situation, there are going to be three different subnets. The connection to Palaestra A is 1.0 subnet, Palaestra B 2.0 subnet and Palaestra C would be the 3.0 subnet. So they actually are three different subnets that the IPs are going to be in. So you got to see IP address for this subinterface 0.1 here of 1.1.
The next CCNA step is mapping the virtual circuit to the subinterface. So each subinterface hosts its own virtual circuit and that’s frame-relay interface-dlci 20. So DLCI of 20 which means ending up at the router with a Global DLCI of 20 is Palaestra A. So it’s saying hey that subinterface that will host that virtual circuit is 0.1. Then you go to serial 0.2, create an additional subinterface, point-to-point. IP address 2.1 this time, again, this is the 0.2 subinterface and mapping DLCI of 30 to it. So again, you are saying, hey, this virtual circuit will host DLCI 30 and interface serial 0.3 point-to-point, 3.1 IP for the appropriate subnet and then map DLCI 40 to it. So Palaestra Central will see three DLCIs 20, 30 and 40 and the next hop router would be A, B and C, Palaestra A, Palaestra B, Palaestra C. To find out the IP addresses of the next hops, inverse ARP would take place, unless we wanted to statically map it using that frame-relay map command.
Let me this clear this off and we will take a look at the configuration of Palaestra A. Here is Palaestra A’s configuration, same thing, interface serial 0, get rid of the IP address, encapsulation frame-relay, serial 0.1 point-to-point, give it the IP address of 1.2 because it’s 1.0 subnet and we are saying, hey, this guy has a global DLCI of 10 for Palaestra Central so you are going to go interface-dlci 10, map this DLCI of 10 to the subinterface 0.1. Same thing for router B and router C. Let’s pop those up real quick. And here is the config, I am drawing a line in between the two of Palaestra B and Palaestra C which are down here. Here is B, here is C, same thing. The only difference is the IP address. And the reason the DLCI is the same is because the Global DLCI of Palaestra Central is 10, that means Palaestra A it looks at DLCI of 10, B looks at DLCI of 10, C looks at DLCI of 10 because that’s where they are headed. So, 2.2 because this is the 2.0 subnet here so this subinterface has an IP of 2.2, and for Palaestra C, 3.2 because this is 3.0 subnet, so this guy has got an IP address of 3.2. And again, frame-relay interface-dlci applies the virtual circuit to the subinterface. Now let’s go, take a look at the last one where we have a Mesh and a point-to-point interface. So one is going to be multipoint, the other one point-to-point.
This is called a Partial Mesh. So we have got serial 0 interface, a global DLCI of 10 for Palaestra Central, 20, 30 and 40 for Palaestra A, B and C. And the big difference here is there is a point-to-point connection between Palaestra Central and Palaestra C. Palaestra Central, Palaestra A, Palaestra B have a Mesh environment, so this is called a Partial Mesh because you have any-to-any connectivity in this area between Central, A and B and then just a point-to-point connection between Central and Palaestra C. So in this situation, we create two subinterfaces, serial 0.1 and serial 0.2, one to host at the multipoint environment and one to host this point-to-point connection. So we need two subnets.
Same thing as before, interface serial 0, encapsulation frame-relay, Enter, no IP and then you go serial 0.1 and you specify multipoint. You are basically telling the router that more than one virtual circuit will be mapped to this interface. Give it an IP. This whole thing right here is the 0.1 subnet so this IP is 1.1 for this subinterface and you are mapping DLCI 20 and DLCI 30 to it. And again, Inverse ARP would tell us that IP address of let’s say this is 1.2 and 1.3 for the appropriate subinterfaces, Inverse ARP would map that information for us. Then you go to the point-to-point interface, serial 0.2, and that’s this one over here, 0.2 and give it an appropriate IP address. This is subnet 2.0 and so this IP is 2.1, this IP 2.2 and then you map the DLCI of 40 to it. So 20 and 30 get mapped to serial 0.1, 40 gets mapped to serial 0.2.
Let’s take a look at the configuration of router A as well as router C and look at the differences. So I brought up the configuration of A and C. I did not bring up the configuration of B because it’s pretty much the same as A with a different IP address. So A right here, this interface, you are going to create a subinterface of serial 0.1 and it’s multipoint because two DLCI virtual circuits will be mapped to that subinterface, virtual circuit with a DLCI of 10 which goes to Palaestra Central and virtual circuit with a DLCI of 30 which goes to Palaestra B. So we look at that interface serial 0, encapsulation frame-relay, no IP, 0.1 multipoint, appropriate IP for the subnet and then frame-relay interface DLCI 10 and interface DLCI 30. So we map those two dlcis to the interface and again, Inverse ARP would allow us to find out that. Interface dlci with 10 takes us to 1.1; Interface DLCI 30 takes us to 1.3. Palaestra C, that’s the point-to-point environment, encapsulation frame-relay, no IP, 0.1, point-to-point. So only one DLCI is going to get mapped to it and the IP address is 2.2 because it’s in the 2.0 subnet. So those are different environments and basic configurations you are going to see for frame-relay.
What I want to do next is I am going to go in and configure frame-relay on my CCNA simulator and then we are going to take a look at some show commands and some trouble shooting. I have gone in and I have configured my CCNA simulator to look like this, Palaestra 1, Palaestra 2, Palaestra 3, have frame-relay connections within one another. Now what I have done is I have gone in and configured in the back to back environment so I have kind of statically mapped all the DLCIs to each other. There are some show commands like Show Run. I love Show Run because it shows me what I have gone in and configured, interface serial 0.1 multipoint, given it an IP address. And I have mapped the next hop IP to a DLCI and I have also associated the DLCIs with the subinterface. So I have gone and configured that. And I can ping right now but there are some show commands that we should be aware of.
The first one let’s say if I try pinging 172.16.1.2 which is Palaestra 2 and it didn’t work, I might want to check out the status of my permanent virtual circuit. So show frame-relay pvc and I can go in and see the status of my virtual circuit. It shows me right here, DLCI – 102, USAGE – LOCAL, PVC STATUS – ACTIVE, that is good. Inactive would be bad, or deleted would be bad. So active is what we want. If it's non-active, I need to go check out my configuration and things like that, see what might be the problem. DLCI – 103, STATUS – ACTIVE, shows me the subinterface, everything like that. It's a very good command. There is also another command that we can use to make sure we have the right IP to DLCI mappings, show frame-relay map. And I can go in and again on the simulator it gets a little bit funky, there is some extra stuff in here but bottom-line is I can go in here and I can see 172.16.1.2 is mapped to DLCI 102. And it shows me that it's statically meaning I am not using Inverse ARP to do it, I have statically mapped the DLCIs to the IP address. In the simulator, it works better that way. And I can see here 172.16.1.3 is mapped to DLCI 103. This is the IP address of Palaestra 3 and I take DLCI 103 which is the virtual circuit with the DLCI 103 to get there. And again, it also shows me the status of the DLCI. It's very helpful.
Another one that I can do is I can debug information if I want to see what’s going on, debug frame-relay lmi, hopefully the CCNA simulator will let me do it. Doesn’t let me do it so write this command down, debug frame-relay lmi, and what that will do is it will allow me to see those lmi status messages going back and forth between my router and the frame-relay service provider switch. So there are some helpful commands, never forget Show Run. And on top of the Show Run, you got the show frame-relay pvc and the show frame-relay map command as well as the debug frame-relay lmi, so very helpful commands when we are troubleshooting why we are unable to ping.
Now let’s go back and do a recap of the commands and a quick recap of what we have gone over. Here is the list of commands and you should definitely be familiar with these commands. Encapsulation frame-relay, if we don’t specify anything Cisco is a type field that’s going to be put in there, ietf for multivendor. Frame-relay lmi-type if we have to specify ansiq933a or Cisco. This is auto-sensed, so we don’t necessarily have to specify it. No frame-relay inverse-ARP, that turns off Inverse ARP. Frame-relay map
Frame-relay interface dlci and the dlci is the command that we use to map those virtual circuits to the interface and then some help. Never forget Show Run by the way. Show running config, I am lazy so I type this, Show Run. And there are some others. Show frame-relay pvc gives us status messages of our permanent virtual circuits, active/inactive. Show frame-relay map will show us the DLCI and the next hop IP. That’s important. I want to make sure those are all correct. Show frame-relay lmi. That one will show us the lmi information, mainly the LMI type, how often we are sending the information out. And then debug frame-relay lmi, so we can see those messages as they happen. So a bunch of very helpful commands when configuring frame-relay, helpful if not necessary.
So we have talked about point-point versus multipoint, the big difference there again, one virtual circuit mapping versus many, what Inverse ARP will do for us as well as how we statically do what Inverse ARP does. We have looked at multiple sample configurations from Partial Mesh, Full Mesh, point-to-point, all those different sample configurations. We went over the commands, we went into the simulator, looked at the configuration of it and went over some troubleshooting commands, all our helpful Show commands. So I hope you have enjoyed this CertificationKits CCNA Training Video on Configuring Frame Relay.