Frame Relay Question
nevolved
Member Posts: 131
in CCNA & CCENT
Ok everyone I have a couple of questions about Frame Relay. I just don't seem to have a 100% grasp on Frame Relay. Let me state some things the way I see them, and have you correct them if they are wrong (please).
Multi-point / Physical interface configuration: requires either Inverse ARP and LMI auto sense to learn the DLCI/IP address combinations, or requires a frame-relay map command
Point-to-Point: does not require inverse arp or lmi auto sense, but requires a frame-relay interface-dlci command to inform the router which dlci to send the frames on
Soooo, why are the requirements different? I can use the above knowledge (if it's right) to pass the test, but WHY are those requirements the way they are. If you configure a physical interface or multipoint you can just have the encapsulation frame command and NOTHING else and be up and running with multiple DLCIs, yet you have to specify which DLCI is on the point-to-point interface and it won't autosense the DLCI/IP address(remote ip) to send on.
I know that the above is rambling, but I really want to get this straight.
Thanks,
And one other thing,
LMI will give you the DLCI information for that PVC, and inverse arp will give you the remote IP address based on the DLCI?
Multi-point / Physical interface configuration: requires either Inverse ARP and LMI auto sense to learn the DLCI/IP address combinations, or requires a frame-relay map command
Point-to-Point: does not require inverse arp or lmi auto sense, but requires a frame-relay interface-dlci command to inform the router which dlci to send the frames on
Soooo, why are the requirements different? I can use the above knowledge (if it's right) to pass the test, but WHY are those requirements the way they are. If you configure a physical interface or multipoint you can just have the encapsulation frame command and NOTHING else and be up and running with multiple DLCIs, yet you have to specify which DLCI is on the point-to-point interface and it won't autosense the DLCI/IP address(remote ip) to send on.
I know that the above is rambling, but I really want to get this straight.
Thanks,
And one other thing,
LMI will give you the DLCI information for that PVC, and inverse arp will give you the remote IP address based on the DLCI?
Comments
-
kalebksp
Member Posts: 1,033 ■■■■■□□□□□
Multi-point / Physical interface configuration: requires either Inverse ARP and LMI auto sense to learn the DLCI/IP address combinations, or requires a frame-relay map command
Point-to-Point: does not require inverse arp or lmi auto sense, but requires a frame-relay interface-dlci command to inform the router which dlci to send the frames on
Soooo, why are the requirements different?
Correct. You must specify the DLCI for point-to-point because otherwise it wouldn't know which DLCI you want to be associated with the sub interface. Point-to-point is useful because you can use it just like you would any serial connection. Multipoint is more difficult since the router thinks it a multi access topology (all routers can reach each other directly), but if you have a hub and spoke topology with multipoint the spoke routers must go through the hub router to reach each other. This causes issues with communication, particularly routing protocols and split horizon and OSPF DRs.LMI will give you the DLCI information for that PVC, and inverse arp will give you the remote IP address based on the DLCI?
Yep.