Frame Relay

whoflungdung88

Question,
I don't wanna do any overkill, nor do I wanna under-study Frame relay for the CCNA Exam. But I have been doing some studying and have noticed that on this site there is one direction on how to setup a frame relay network, and on other study resources I have been using show either a slightly different setup, or just a different way to configure frame relay.
So I ask, what kind or kinds of frame relay configurations should I concentrate on? Or can somebody point me to a reference of all the different configuartion scenarious that I would have to know how to do? Thanks

Netstudent

I would master Full-Mesh all interfaces in 1 subnet, Partial-mesh (point-to-point subinterfaces with each PVC on it's own subnet) and Hybrid(which i think requires the (config)# int s0/0.1 multipoint command).

Now with full mesh, I would master both ways. Inverse-Arp which will create the mappings by creating the routes in the frame switch. the LMI's will tell the routers whos who as far as layer 2 and 3 mappings.

And then static mappings using the Frame-relay map command on the routers. If you can master those implementations, in my opinion, you should be good to go.

I would nail it down, be able to do each of those ad-hoc. Right off the top of your head.
That way you can do some serious troubleshooting if the need arises.

One thing I will add that I just recently understood is the difference in Inverse-ARP, and static mappings with reference to GLOBAL DLCI's.

you will probably see this as well as you go deeper into frame-relay. Lets say you have 3 end points like in the techlabs example. This is hard to explain but, with the inverse-ARP method, where you create the routes in the switch, you might use a total of 6 DLCI's for 3 Virtual circuits. Like in the example in techlabs.


BUT! with the static mapping method you can use a total of only 3 dlci's for 3 virtual circuits! WHY? well because DLCI's only need local significance. What I;m trying to explain may not be clicking because I'm not good at explaining it. But look at the example in techlabs and then look at an example in a CCNA book that uses static mappings. You may have already came to this realization. I just recently did a couple days ago and it was like a frame-relay epiphany. It all just came togther right then. I'm still learning too!

One thing I haven't figured out on my lab, is why my OSPF isn't creating adjacencies across a NBMA network like frame-relay. I'm gonna tackle it this evening and if I don;t figure it out, I'll be posting my own thread about it.

AND remember OVERKILL IS GOOD!!!! What most people probably see as overkill, is what really needs to happen to pass this test the 1st time. Employers want knowledge, not the letters CCNA on a resume. Only HR cares about that.

Netstudent

One other thing, I haven't read Cisco press or Lammle(Sybex delux) explain frame-relay switching at all. The only material that I have seen in both of these books is all about end point router configurations . HMMM...I wonder what this could mean.....

And I find this strange since SOME frame-relay configurations will not bring up a PVC without a frame-relay route created in the switch.