Multicast(Basic Question)

Hi everyone,

i want to ask a basic question about multicasting, if a router(e.g R1) has 2 path(e.g 2 interfaces fa0/0 and fa0/1) to reach RP(e.g R3) and both of them enabled for multicast(ip pim sparse), and R1 has receive an IGMP join message(e.g on interface fa1/0) then how R1 will choose the interface that should sent join message over it? and will mroute affect this selection ?

 

Regards

Comments

  • Hi,

    It will choose base on your CEF table, so whatever is the outgoing preferred interface for that pair will be chosen. If you configure a mroute it will definitely affect the decision and override with what you statically configured.

    HTH

    Good luck!

  • R1 will choose the interface that should sent join message over it? and will mroute affect this selection

    Assuming that no mroutes are configured then the join will be sent out the closest interface to the the RP.  What I mean my closest is the lowest metric - you can see this by doing show ip route <rp-address> . Clearly PIM must be running on this interface otherwise you will get into all sorts of problems a.k.a reverse path forwarding failures (RPF). You can also use the command show ip rpf for more information.

    Reverse Path Forwading is the key to home multicast control plane signalling works.  Like most things in multicast this mechanism is completely counter intiative in comparision to unicast routing.

    Let me know if you need anymore information. 

     

  • Thanks all i appreciate yours help

  • i want to ask a basic question about multicasting, if a router(e.g R1) has 2 path(e.g 2 interfaces fa0/0 and fa0/1) to reach RP(e.g R3) and both of them enabled for multicast(ip pim sparse), and R1 has receive an IGMP join message(e.g on interface fa1/0) then how R1 will choose the interface that should sent join message over it? and will mroute affect this selection ?

    Also think of a scenario where RP is closest via f1/0 (where igmp join message is received).

     

  • In section 5.3 i dont understand why their wouldnt be a RPF failure going to the vlan 52?  Doesnt the IGP route take the FA link between R5 and R4.  I am pretty confused on 5.3.

  •  

    Hi,

    Before switching to the SPT, the RPF check is performed against the RP, so there is no failure in this state. After switching to the SPT, the check is performed against the source. In this state, the traffic follows the path you mentioned, and again there is no failure.

     

    HTH,

    Bassam

  • Bassam, thanks.  it will flow the FA link after the SPT switchover even though PIM is not configured on those interfaces?

     

    The RP should be R1 right? Since it is for even groups?

    I will have to check my lab tonight, i thought i followed the SG and couldnt get it to work.

     

  •  

    OK, then switching to the SPT should fail. I'm not sure of the following information, which I have understood from Brain M. in the lab preparation bootcamp: "Another way to stay on the RPT (i.e. not switched to the SPT), is to cause the join to the SPT to fail." Can any one please confirm this?!

  • You can use the ip pim spt-threshold infinity so it never tries to switch over to SPT.

    HTH

  • You can use the ip pim spt-threshold infinity so it never tries to switch over to SPT.

     

    You are right for sure. But what I mean is that in one of scenarios Brain was working on, a router had to switch to SPT ( so there was no ip pim spt-threshold infinity), and when that failed, it just remained on the RPT, and everything was fine. I need someone to confirm this, and if that  technique is valid for solving RPF failures. That is my point!

    Regards!

     

  •  

    Hi mmokeev,

    After checking lab 8 solutions, I think the SPT switchover will succeed over the FA interface. In task 2.5, EIGRP load balancing was configured, so that R4 has two routes to the 192.10.x.0 network, one through R1, and the other through R5 directly. However, the RPF ckeck should result in only one route, which should be through the frame-relay link because the "active PIM neighbor" is there. I found the following in my own notes:

    " If router has multiple equal-cost paths to the source. Firstly, only those routes that point to the active PIM neighbors would be used. Secondly, the router will use the entry with the highest PIM neighbor IP address."

    This was written by Petr Labukhov here: Understanding static multicast routes

    It is great and very useful post by Petr as usual.

     

    HTH,

    Bassam

Sign In or Register to comment.