Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 
New Member

Bidirectional PIM

Hi all,

I need some clarification about the following point. Let's consider a Bidirectional PIM deployment and, more specifically, a shared tree branch without receivers, that is a branch to which only multicast sources are attached. Will there be (*,G) state in the routers forming this branch? And, if yes, how is it formed, considering that no PIM Join is tranmitted through such routers?

I have looked up RC 5015 and it states:

"Source-only branches of the distribution tree for a group G are    branches that do not lead to any receivers, but that are used to    forward packets traveling upstream from sources towards the RPL.    Routers along source-only branches only have the RPF interface to the    RPA in their olist for G, and hence do not need to maintain any group       specific state.  Upstream forwarding can be performed using only RPA    specific state.  An implementation may decide to maintain group state    for source-only branches for accounting or performance reasons.    However, doing so requires data-driven events (to discover the groups    with active sources), thus sacrificing one of the main benefits of    BIDIR-PIM."

While in Cisco documentation about Bidir PIM I have found that "If a router is located on a sender-only branch, it will also create (*, G) state, but the olist will not include any interfaces"

I'm bit confused...Could someone help me out please? 


Bidirectional PIM


Source traffic follows the shared tree

  going up the Shared Tree to reach the RP, in the OIL will be as Bidir-Upstream

  going down the Shared Tree to reach all other receivers, in OIL will be as Forward

If there are no connected receiveres, in the OIL will be only the interface toward the RP.

The creation of the Shared tree is the same , based on PIM join/prune messages.

Other differences :

    - No Source registration.

    - No SPT switchover


CreatePlease to create content