Frame-relay subinterface & split-horizon

[news2010a]

Hi, I have been reading about this and I would like to confirm my understanding is correct:

Imagine I setup a R1-Hub router, s0/0 physical interface connected to (3) spoke routers R2,R3,R4 like shown below.

Questions:

1) If I am running a distance vector protocol, is the split horizon behavior going to apply to the multipoint subinterface below? Note all spokes are on the same subnet... so it seems to me split horizon would still be in effect given the config below.

I know in point-to-point subinterfaces split horizon does not apply.

#R1

(...)

int s0/0

   encap frame-relay

    no ip address

    no shut

    no frame-relay inverse-arp

int s0/0.234 poin-to-multipoint

     ip add 192.168.1.1 255.255.255.0

     frame-relay map ip 192.168.1.2 122 broad

     frame-relay map ip 192.168.1.3 133 broad

     frame-relay map ip 192.168.1.4 144 broad

(...)

[Richard Burts]

Your configuration would result in split horizon issues. If the routing protocol you are running does enforce split horizon then R1 will learn an update from R2 but will not forward the update to R3 or to R4. The result is that the hub knows how to get to all subnets, but each spoke knows only its local subnet and the subnets from the hub.

HTH

Rick

[cadet alain]

Hi,

yes in RIPv2 split-horizon is enabled on frame-relay multipoint subinterfaces(but disabled on physical interfaces) and in EIGRP it is enabled for both.

Regards

Alain

Don't forget to rate helpful posts.