RE: [pim] ABOUT Update_SPTbit(S,G,iif)

["Nataraja, Rajesh \(Rajesh\)" <rnataraja at lucent.com>]

Title: Message

Hi  chunyan,

 

This check is necessary for the case where the router which has to trigger SG RPT prune, has the IIF towards the RP and the IIF towards the Source to be the same. But the RPF neighbors towards them are different.

In this case there is no way the router can tell if the packets are arriving on the SPT or RPT. Only the Assert from the SPT-neighboring router which wins against the RPT-neighboring router can explain if the SPT switch has happened.

 

 

Thanks

Rajesh.

 

---

From: CTO YAO Chunyan [mailto:Chunyan.YAO at alcatel-sbell.com.cn]
Sent: Monday, November 27, 2006 4:43 PM
To: pim at ietf.org
Subject: [pim] ABOUT Update_SPTbit(S,G,iif)

 

Dear all, SPTbit is updated as follows:

     void
     Update_SPTbit(S,G,iif) {
       if ( iif == RPF_interface(S)
             AND JoinDesired(S,G) == TRUE
             AND ( DirectlyConnected(S) == TRUE
                   OR RPF_interface(S) != RPF_interface(RP(G))
                   OR inherited_olist(S,G,rpt) == NULL
                   OR ( ( RPF'(S,G) == RPF'(*,G) ) AND
                        ( RPF'(S,G) != NULL ) )
                   OR ( I_Am_Assert_Loser(S,G,iif) ) {
          Set SPTbit(S,G) to TRUE
       }
     }
 

 

I don't understand HOW the last one condition " I_Am_Assert_Loser(S,G,iif)" along with "iif == RPF_interface(S)AND JoinDesired(S,G) == TRUE" can show that source-specific-tree is OK?

 

Could you tell me what the condition " I_Am_Assert_Loser(S,G,iif)"mean?

 

 

Thanks a lot in advance!

 

                       Chunyan   

_______________________________________________
pim mailing list
pim at ietf.org
https://www1.ietf.org/mailman/listinfo/pim