Reading Time: 5 minutes
Editor’s Note: Google Chrome seems to dislike my site theme and is hyphenating absolutely everything. Apologies for that, and I’ll look into it just as soon as I get done with a few items on the “honey-do” list.
If you are studying for the CCIE Routing and Switching exam, one of the technologies that is still heavily prevalent is Frame Relay. It is expected that you know both the technology itself, and how to configure it, but also how it interacts with and affects other key technologies like OSPF and EIGRP. Having the ability to study Frame Relay, then, and get plenty of hands-on configuration time becomes as important as with anything on the R&S 4.0 Blueprint.
While many network engineers are already familiar with Frame Relay from a consumer side–in other words, from the perspective of an entity which buys Frame Relay services from a provider–not many of us are familiar with the service provider portion of the equation. This makes setting up practice labs difficult if you are trying to study using your own equipment. Fortunately, you can set up your own Frame Relay switch fairly easily, and that is what we’re going to walk through today.
A Frame Relay switch is the DCE device that sits inside a service provider’s network and moves the frames along from point A to point B. There are many of these devices all working together inside of your provider’s network to move your information along, but fortunately for lab candidates studying at home, you can easily get by with just one. Even more fortunate is that you can use a fairly low powered router to act as a Frame Relay switch, and not miss anything that you’ll need for purposes of the lab.
A quick note on the lab is in order here. It used to be a part of the lab blueprint (dont’ ask me which one, or how far back in time) that you had to know how to set up a Frame Relay switch. Cisco has since taken that requirement away, at least from the R&S lab, and so a lot of that knowledge isn’t communicated in teaching texts any longer. What you’ll find in the lab itself is an already configured Frame Relay switch that you’ll have no direct access to, but all of the information you need to make your equipment talk to it.
It may seem counterintuitive, but for a home lab the best device to use for a Frame Switch is actually a router. For instance, I’m using an older Cisco 2621 model for my Frame Switch, and it does everything I need it to do. Service providers will typically use more specialized gear, but all we’re going for in our studies is a reasonable facsimile. If you want to spend a lot of money, follow the advice of so many others and spend it on your layer‑3 switches.
Another thing we want to briefly discuss is interfaces. Generally speaking, you can either follow the “run what’cha brung” philosophy of just using what you have access to, or you can buy the interfaces you want. In my case I had a couple of WIC-T1 cards that I’ve used, and then I bought a handful of WIC-2T serial interface cards. The key is to have a serial interface for each router you want to connect via the Frame switch. So I have one T1 interface, and six serial interfaces for a total of seven devices I can connect into the Frame “cloud”. I find this to me more than adequate, though if you’re trying to duplicate a specific topology you may need more or less.
The configuration of a Frame switch is actually very simple, as you’ll see, though attention to detail does matter. I’m assuming here, by the way, that you already know how to set up your router for basic access, clock, etc., so I won’t cover that here. So, the first step in configuring your router to be a Frame switch is to put it into Frame switching mode using the commands:
ip cef
frame-relay switching
These commands turn on Cisco Express Forwarding, put the router into a Frame switching mode, and change quite a bit of the default behavior, so don’t expect to use this device as a router in any lab topology you’re working on. This device will be just a Frame switch and nothing more.
The next step is to configure the individual interfaces you’ll connect your other routers to, and you have a lot of choices here. I don’t know exactly how the R&S lab devices are set up, so I’m just going to give you the configuration I use. I’ll post the configuration below, and then go over the key comands:
interface Serial0/1
no ip address
encapsulation frame-relay
logging event subif-link-status
logging event dlci-status-change
clock rate 8000000
no frame-relay inverse-arp
frame-relay intf-type dce
frame-relay route 220 interface Serial0/2 120
frame-relay route 221 interface Serial0/0 320
The first few lines of the configuration should be familiar to you already. We’re setting our interface encapsulation to frame-relay, and then logging on a couple of events. The logging is completely up to you, and not necessary one way or another. I just find them helpful. Next we set the clock rate, and we tell the interface that we are the DCE end of the connection. Remember, in a Frame Relay network the clocking (DCE end) comes from the line or provider side, so this is what you’ll want. If I am working with a T1 serial interface, I’ll also need a line for that:
service-module t1 clock source internal
This can change depending on the type of card and how you have it configured.
Now, the other options we have here require a little more explanation. The “no frame-relay inverse-arp” command does just what it says, and you can argue for the Frame switch having this turned on, or off. In most cases in the lab, you’ll be instructed to not use inverse arp on the DTE devices, so I’ve just turned that functionality off on my Frame switch from the outset. It’s really your call.
The next two lines beginning with frame-relay route are the ones that always seem to cause confusion. You can read the first line as “If some traffic comes in from DLCI 220, with a destination of DLCI 120, send it out interface Serial 0/2”. Substitute DLCI 221 and 320 on the next line, but otherwise read it the same. So if I now plug in a router to interface Serial 0/1, and assign DLCI 220 and 221 to two different sub-interfaces (for instance, different options are possible) the Frame switch will know what to do with that traffic.
So, if we have a diagram that looks like the following:
Then we have a configuration for interfaces that looks like so:
interface Serial0/0
no ip address
encapsulation frame-relay
logging event subif-link-status
logging event dlci-status-change
service-module t1 clock source internal
no frame-relay inverse-arp
frame-relay intf-type dce
frame-relay route 320 interface Serial0/1 221
frame-relay route 321 interface Serial0/2 121
!
interface Serial0/1
no ip address
encapsulation frame-relay
logging event subif-link-status
logging event dlci-status-change
clock rate 8000000
no frame-relay inverse-arp
frame-relay intf-type dce
frame-relay route 220 interface Serial0/2 120
frame-relay route 221 interface Serial0/0 320
!
interface Serial0/2
no ip address
encapsulation frame-relay
logging event subif-link-status
logging event dlci-status-change
clock rate 8000000
no frame-relay inverse-arp
frame-relay intf-type dce
frame-relay route 120 interface Serial0/1 220
frame-relay route 121 interface Serial0/0 321
I hope that helps out, and as always if you have any questions or clarifications please drop me a line here or on twitter where I’m known as @SomeClown.