[j-nsp] R: Re: R: Re: Same multicast group from the source

[chris at chrishellberg.com]

Almost right. The bit that ssm mapping does is take a igmpv2 group reports and maps them into a PIM-SSM towards the source(s). Thus no need for an RP....as if the request were igmpv3

Been years since I've looked at this but from memory you don't even need to do ssm mapping since the rp should join both sources and you'll get a multiplexed stream, which is what you want. If I understand correctly.

/chris
---

-----Original Message-----
From: Riccardo S <dim0sal at hotmail.com>
Date: Thu, 22 Nov 2012 21:47:24 
To: <chris at chrishellberg.com>; Stacy W. Smith<stacy at acm.org>
Cc: <juniper-nsp at puck.nether.net>
Subject: R: Re: R: Re: [j-nsp] Same multicast group from the source

As far as i know ssm works with igmp3
With igmp2 you need to go through rp...

sent with Android

chris at chrishellberg.com ha scritto:

>Yes it does
>---
>
>-----Original Message-----
>From: Riccardo S <dim0sal at hotmail.com>
>Date: Thu, 22 Nov 2012 20:56:43 
>To: Stacy W. Smith<stacy at acm.org>
>Cc: <chris at chrishellberg.com>; <juniper-nsp at puck.nether.net>
>Subject: R: Re: [j-nsp] Same multicast group from the source
>
>I cannot force my customer to use igmpv3....
>Do not think it works with igmpv2
>
>sent with Android
>
>"Stacy W. Smith" <stacy at acm.org> ha scritto:
>
>>You might want to look at ssm-map-policy.
>>
>>http://www.juniper.net/techpubs/en_US/junos12.2/topics/topic-map/multicast-ssm-map-for-different-groups-to-different-sources.html
>>
>>I think it meets your needs, but depending on how dynamic your customer's requirements are, it may be cumbersome to maintain.
>>
>>Here's an example:
>>
>>[edit]
>>user at host# show protocols igmp     
>>interface ge-0/0/0.0 {
>>    ssm-map-policy S1-only;
>>}
>>interface ge-0/0/1.0 {
>>    ssm-map-policy S1-and-S2;
>>}
>>interface ge-0/0/2.0 {
>>    ssm-map-policy S2-only;
>>}
>>
>>[edit]
>>user at host# show policy-options 
>>policy-statement S1-and-S2 {
>>    term 1 {
>>        from {
>>            route-filter 224.1.1.1/32 exact;
>>        }
>>        then {
>>            ssm-source [ 1.1.1.1 2.2.2.2 ];
>>            accept;
>>        }
>>    }
>>}
>>policy-statement S1-only {
>>    term 1 {
>>        from {
>>            route-filter 224.1.1.1/32 exact;
>>        }
>>        then {
>>            ssm-source 1.1.1.1;
>>            accept;
>>        }
>>    }
>>}
>>policy-statement S2-only {
>>    from {                              
>>        route-filter 224.1.1.1/32 exact;
>>    }
>>    then {
>>        ssm-source 2.2.2.2;
>>        accept;
>>    }
>>}
>>
>>
>>
>>On Nov 22, 2012, at 9:54 AM, Riccardo S <dim0sal at hotmail.com> wrote:
>>> 
>>> no failover, are completely different data...
>>> 
>>> Let's imagine we're talking of TV broadcasting and my customer wants receive today only channel S1 but tomorrow also, maybe, channel S2...
>>> 
>>> Ric
>>> 
>>>> Date: Thu, 22 Nov 2012 17:51:07 +0100
>>>> From: chris at chrishellberg.com
>>>> To: dim0sal at hotmail.com
>>>> CC: juniper-nsp at puck.nether.net
>>>> Subject: Re: [j-nsp] Same multicast group from the source
>>>> 
>>>> It depends on what you want to do. Should the receiver receive both 
>>>> sources simultaneously and demultiplex the two? Or do you need failover 
>>>> of some kind?
>>>> 
>>>> /Chris
>>>> 
>>>> On 22/11/12 5:40 PM, Riccardo S wrote:
>>>>> 
>>>>> 
>>>>> 
>>>>> Hi expert team,
>>>>> 
>>>>> I’ve a generic multicast question.
>>>>> 
>>>>> 
>>>>> 
>>>>> Let’s say I’ve two different source
>>>>> sending different multicast traffic over the same multicast group, how can I
>>>>> solve the problem for a customer receiver ?
>>>>> 
>>>>> 
>>>>> 
>>>>> S1=1.1.1.1 G1=224.1.1.1
>>>>> 
>>>>> S2=2.2.2.2 G2=224.1.1.1
>>>>> 
>>>>> 
>>>>> 
>>>>> Customer receiver is usually running
>>>>> IGMPv2 and I cannot force to use IGMPv3.
>>>>> 
>>>>> 
>>>>> 
>>>>> Network in the middle between source
>>>>> and receiver is not MPLS.
>>>>> 
>>>>> 
>>>>> 
>>>>> I’ve thought to NAT multicast group,
>>>>> but I’m not sure it works and if anyone has already implemented it…
>>>>> 
>>>>> 
>>>>> 
>>>>> Any different solution is
>>>>> appreciated…
>>>>> 
>>>>> Tks
>>>>> 
>>>>> Ric
>>>>>  		 	   		
>>>>> _______________________________________________
>>>>> juniper-nsp mailing list juniper-nsp at puck.nether.net
>>>>> https://puck.nether.net/mailman/listinfo/juniper-nsp
>>>>> 
>>> 		 	   		  
>>> _______________________________________________
>>> juniper-nsp mailing list juniper-nsp at puck.nether.net
>>> https://puck.nether.net/mailman/listinfo/juniper-nsp
>>
>>