Re: Multicast issue in NBMA

Have you enabled "ip pim nbma-mode" on the hub? I'd imagine we need to get past the split-horizon like behaviour in order to register with the RP, the RP will then be cut out of the path due to SPT optimisation depending on where you issue the pings from... another alternative is to create a GRE tunnel between source and RP to make them think their connected directly.

--
BR
Tony
Sent from my iPhone on 3
On 4 Jul 2013, at 18:43, Taio Bone <taiobtnh_at_gmail.com> wrote:
> Hello expert. I'm having a challenge with a certain scenario. Say we have 3
> routers(R1,R2 and R3) with R1 as the hub (in a frame relay network with
> ospf  network type "BROADCAST") while R2 and R3 are spoke routers.
> 
> All interfaces are in sparse mode. Now,I have a host behind R2 trying to
> join a multicast group whose RP is on the loopback(3.3.3.3/32) of the other
> spoke router,R3.
> 
> Now,the issue I'm encountering is that R2 gets the join message and has an
> entry for (*,G) but R1 isn't getting the join message neither does it have
> an entry for (*,G).
> 
> You assume all the routers are connected to the FRS using their Serial1/0
> with IP addresses 10.10.10.(1,2,3) for R1,R2 and R3 respectively.
> 
> What I'm assuming might be the problem is that fact that the next hop and
> RPF neighbor indicated on R2 for the RP address  is 10.10.10.3..Now this is
> expected since it's an OSPF "broadcast" network type and they are all in
> the same broadcast domain.
> 
> P.S Each spoke router has a frame-relay map statements to both the hub and
> the spoke routers.
> 
> Thank you.
> 
> 
> 
> 
> -- 
> Warm wishes,
> 
> Ty.
> 
> 
> Blogs and organic groups at http://www.ccie.net
> 
> _______________________________________________________________________
> Subscription information may be found at: 
> http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net