All things related to multicast

[Peter John Hill <>]

If you are using Auto-RP you need to use either [dense-mode or 
sparse-dense-mode] or anycast-rp

The idea is that the leaf routers need a way to discover the RP. With 
anycast RP, you have the leaf routers use special multicast groups 
(224.0.1.39 and 224.0.1.40) to discover the real rps. It works fine. We 
have two routers that are msdp peers that are both configured with the 
same loopback1 address of 128.2.1.130. This ip is distributed 
throughout OSPF to all the leaf routers. The leaf routers find the 
closest rp and then that rp hands out their unique ip address that the 
leaf routers use as the rp address. Because we have an msdp mesh, both 
of the rps have the same msdp state. If one rp dies, then the leaf 
routers will use the other rp to setup new multicast sessions. If you 
look at the link below, you can see that of all the different pim 
methods, that auto-rp, anycast, sparse mode is the way to go.


On Wednesday, September 4, 2002, at 12:13  PM, Tim Ward wrote:

> Quoting Peter John Hill 
> ():
> > Totally been there and done that. We just got everything stable a few
> > weeks ago. Give me a call if you want to discuss it
> ...
> > We are using auto-rp with anycast and sparse mode. It is #10 as
> > described in:
> > http://www.cisco.com/warp/public/cc/pd/iosw/tech/rppim_rg.htm
>
> When we were rolling out multicast to all of our dorms for our dorm
> TV system, we were seeing odd 60 second dropouts.  We had at the time
> Auto-RP and PIM Sparse.  The scoop from the TAC on the problem is that
> Auto-RP relied on PIM Dense for the auto-RP to work, and that you had
> to configure for PIM Sparse Dense mode if you were using auto-rp.
> (the heavily distilled version).
>
> Does this jive with everyone else's experience?
>
> We were halfway to PIM-Sparse BSR and migrating to anycast rp so I
> didn't think much of it at the time.  But looking at the comments above
> made me wonder.