This topic is to discuss the following lesson:
If we use distance vector routing protocol the next hop ip address will be the hub router in both of broadcast & non-broadcast network types ? that’s right ? or I’am wrong ?
If we use OSPF non-broadcast network type we need to configure spoke router to become OSPF neighbors as well as create additional frame-relay maps ? that’s right ? please correct me if I wrong ?
As long as the spoke routers are neighbors with the hub then the spokes will be able to learn all prefixes. The only problem is that with the broadcast and non-broadcast network types, the spoke won’t know how to reach the next hop of another spoke…that’s what you need the frame-relay maps for.
Routing protocols like EIGRP or RIP don’t have the broadcast and non-broadcast network types that OSPF has. By default they use multicast and if you use the neighbor command they switch to unicast traffic only, that’s it.
The next hop will be the Hub router though, you can see see an example here:
So how can the spoke router reach other with one hop if they did not have PVC ?
What do you mean exactly Hussein?
when I said broadcast & non-broadcast network types I mean if the connection of frame relay was multipoint the network can be broadcast or non-broadcast, so if it was broadcast there is no problem else we have to configure static neighbors, and my question was what is the next hop IP address in in both cases ( broadcast & non-broadcast ), and depending on what you said, I understand that you mean in distance vector routing protocol if we use multicast or unicast the next hop IP address will be always the hub router ? that’s right ?
We have only 2 PVCs :-
1 – between hub & spoke 1.
2 – between nub & spoke 2.
that mean if spoke 1 wants to reach spoke 2 he must go through the hub router ? right ?
My question is if the OSPF network type was “broadcast” or “non-broadcast” the next hop will be the router advertising the network so how the spoke routers can reach each other if there is no PVC between them ? do you mean if we create the frame-relay maps we also create PVC between spoke routers ?
With a hub and spoke topology we only have two PVCs so spoke-to-spoke traffic always has to go through the hub. When you use broadcast or non-broadcast, each spoke will see the next hop IP address of another spoke for updates that they advertise.
In this case, you just create a frame-relay map on each spoke with the DLCI number of the local spoke and the IP address of the remote spoke. The IP packet will be sent to the hub which forwards it to the correct spoke, that’s it.
Ah ok I see, yes you are right. With OSPF the next hop changes depending on which OSPF network type you use.
When you use RIP or EIGRP, the next hop will be the hub, doesn’t matter if you use unicast or multicast for the neighbor adjacencies.
Good Day Rene,
Thank You again for Brilliant explanation. Now I have a challenge to on my lab environment to configure the switch which will connect between HUB and Spoke1 and 2 as frame relay. is it possible to get clarification on that step by step with picture example if possible.
Thank You again.
Take a look here, it’s a step-by-step walkthrough how to configure a router as frame-relay switch:
This is by the far the most explanation for CCNP.
My question is why do we have to elect DR/BDR on Non Broadcast network? I thought DR/BDR are only needed on broadcast networks?
Specifically, a DR/BDR are needed in a multiaccess network. Now some multiaccess networks support broadcast and some do not. Ethernet is a multiaccess network that supports broadcast while Frame-Relay can be configured as a multiaccess network, but does not support broadcast.
Because EIGRP uses broadcast and multicast communication to allow routers to exchange information, the type of network must be known so that it can be configured to function correctly even if broadcast/multicast functionalities are not available.
I hope this has been helpful!