I found a descreptancy and I think its just a typo but I want to confirm that before I commit it to fact.
the post stated the following:
"Frame-relay can use point-to-point sub-interfaces or point-to-multipoint sub-interfaces. If you use point-to-point it will solve your split-horizon problem but you’ll need to use a different IP subnet per PVC.
Point-to-multipoint means you have the split-horizon problem
but you can use a single IP subnet for all PVCs."
I thought the DV rule for split horizon stated it will not send out subnets its already learned on the same interface it received them.
So if you used point to multipoint and set it up with sub interfaces but had all of them on the same subnet. that work around the split horizon problem as well because you might have a sub interface with 192.168.1.1 and 192.168.1.2. It could learn it on .1 and then send it back out on .2 without worrying about split horizon.
is that not correct?
So in that case it would seem point-multipoint with subinterfaces would not have split horizon issue.
or maybe it would be receiving it on both .1 and .2 and so would not send back out. However I don’t think that would happen if you configured it like so as it would have a sub interface for each DLCI and PVC which is like point-2-point but not fully as it would still be on the one subnet.
I understand your logic when you say “So if you used point to multipoint and set it up with sub interfaces but had all of them on the same subnet. that work around the split horizon problem…”
However, a router cannot have two interfaces on the same subnet. Each interface MUST be on a different subnet. This goes for subinterfaces as well. If you try to configure subinterfaces with IP addresses 192.168.1.1/24 and 192.168.1.2/24, you will get an error message when configuring the second interface.
So the following statement stands:
Frame-relay can use point-to-point sub-interfaces or point-to-multipoint sub-interfaces. If you use point-to-point it will solve your split-horizon problem but you’ll need to use a different IP subnet per PVC. Point-to-multipoint means you have the split-horizon problem but you can use a single IP subnet for all PVCs.
I actually did that in my GNS3 lab and did not get an error but one of the sub interfaces was not working and that may have been the problem that while GNSS3 did not give me and error on the c3725 model router I chose that it was malfunctioning in some way. I did not realize that you could not have different subinterfacess with same subnet on the same router setup.
Hub(config)#interface serial 0/0.123 multipoint
Hub(config-subif)#ip address 192.168.123.1 255.255.255.0
Hub(config-subif)#frame-relay map ip 192.168.123.2 102 broadcast
Hub(config-subif)#frame-relay map ip 192.168.123.3 103 broadcast
mistakes are ok but when a new person learning theory and facts its important to concrete ideas and discreptancies can be confusing so just to confirm your saying what was done there is incorrect that you cannot put both of those IPs on sub interfaces as they are in the same subnet and that goes against what you just stated?
Hub(config)#interface serial 0/0.123 multipoint
Hub(config-subif)#ip address 192.168.123.1 255.255.255.0
Hub(config-subif)#frame-relay map ip 192.168.123.2 102 broadcast
Hub(config-subif)#frame-relay map ip 192.168.123.3 103 broadcast
mistakes are ok but when a new person learning theory and facts its important to concrete ideas and discrepancies can be confusing so just to confirm your saying what was done there is incorrect that you cannot put both of those IPs on sub interfaces as they are in the same subnet and that goes against what you just stated?
or never mind that looks like its one subnet configured.
Its two mappings that he has which is different. Ok that does not go against what you said. I was just looking at the mapping to quickly and took that as subnets, my mistake.
I have been working on frame relay all day and I think I pretty much have everything down for it.
I just need to separate when and why to use the following commands listed below better as that last part is a bit blurry in my mind as everything comes together.
frame-relay map ip x.x.x.x DLCI# broadcast
frame-relay interface-dlci DLCI#
I know that number 2 is used with point to point more
also that number 1 is used with multipoint
Questions is:
how can I tell for sure which one will be used. I think if I can put hard decision on that then I am done with frame relay and can move on (at least that’s the last question I can think of at this time lol)
Ok I think I found the answer in a form I can understand I will paste the information below and the link I got it from:
For point-to-point subinterfaces, the destination is presumed to be known and is identified or implied in the frame-relay interface-dlci command. For multipoint subinterfaces, the destinations can be dynamically resolved through the use of Frame Relay Inverse ARP or can be statically mapped through the use of the frame-relay map command.
We know that on Frame Relay relies on the mapping between DLCI to IP Address and this is done by using Inverse ARP. We also know that without using dynamic mapping, we can also use static mapping using command frame-relay interface-dlci or frame-relay map.
The different between the two is that frame-relay interface-dlci is used for point-to-point subinterfaces and frame-relay mapis used for multipoint subinterfaces.
The logic behind it is basically in point-to-point network the next-hop is always known and whatever the interface thrown at will be passed to the next-hop. All we need to do is to define what the interface DLCI number is.
For multipoint subinterfaces, the situation is different. One subinterface faces multiple end-points. Therefore, there is a need to define which DLCI mapped to which IP Address.
It’s great that you were able to clarify your own questions before I got a chance to respond. I believe that when you find the answers in this way, it is extremely beneficial in learning.
I wish you all success in your endeavours, and along the way, we are always here to help!