Re: SSM

I got it to work!

Here is the setup:

R4 -(1.1.24.0)- R2 -(1.1.12.0)- R1 -(1.1.13.0)- R3---loop0 (3.3.3.3)

R4 - 1.1.24.4 pinging to 232.1.1.1
R4 has no multicast, just used to ping 232.1.1.1

R3
int Loop0
ip address 3.3.3.3 /32
ip igmp join-group 232.1.1.1 source 1.1.24.4

R2, R1, R3
ip multicast-routing
ip ssm default
ip pim sparse (all interfaces)
Eigrp 1:
network 0.0.0.0
no auto-summary

Mistake #1 I made was to initially use 239.1.1.1.
For SSM, I have to use 232.0.0.0/8 range! (I found in the doc >.< )
Once I switch to 232 range, I started getting a (S,G), but the (S,G)
is propagated by R3 to R1 to R2 from the igmp join on R3's loop0.
I could see this in the debug.

Mistake #2 I assumed the RPF was correct because it was a no brainer
topology.
The RPF check on R2 had to be failing because he had the source as locally
connected. (I guessed...)
And, the (S,G) was being advertised from R1 to R2 in the wrong interface
from R2's perspective.
The "show ip RPF" was correctly showing the attached interface for 1.1.24.4,
but it apparently needed to point back to R1.
What resolved the issue was that I used a static mroute on R2:

ip mroute 1.1.24.4 255.255.255.255 s0/0 (pointing it toward R1)

But, that doesnt explain why the RPF on R1 doesnt fail.
R1 had the proper incoming/outgoing interface.

The static mroute on R2 ultimately resolved the issue, but I cant think of a
way to resolve the RPF dynamically.
Maybe use a /32 static to null0 on R3 and redistribute that into EIGRP?
(More specific route)
So, I take form this experiment that if you use an igmp join with SSM you
will have an automatic RPF failure at the source.

Chris

On Mon, Sep 20, 2010 at 4:39 PM, Narbik Kocharians <narbikk_at_gmail.com>wrote:

> I think it all Depends on the IOS you are using. I had to configure the
> router that is emulating a host (Receiver) with "NO ip routing" and
> configure "IP Pim SSM range" and NOT "IP Pim SSM default" and of course the
> range references an access-list, and in the access-list i had to permit the
> SSM default range, because i found that the "default" keyword did not work
> on my IOS. But in the lab, i don't think that you will be penalised based on
> IOS bugs.
>
>
>
> On Mon, Sep 20, 2010 at 1:48 PM, Chris Grammer <cgrammer_at_essilorusa.com>wrote:
>
>> Is there any type of method to simulate SSM?
>> According to the doc cd, both the source and receiver must support IGMP
>> version 3.
>> The workbook I am using does give a method to test correct configuration.
>> The only information on the doc cd is simply to configure:
>>
>> ip multicast-routing
>> ip pim ssm default
>> and then on the interface either:
>> ip igmp version 3 or 3lite
>>
>> Just seems kind of pointless to turn something on that cant be verified,
>> tested...etc.
>> This must be one of those type questions that will be something
>> like...turn
>> on SSM and make int fa0/0 igmp version 3?
>>
>>
>> Thanks,
>>
>> Chris
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>>
>>
>>
>>
>>
>>
>
>
> --
> Narbik Kocharians
> CCSI#30832, CCIE# 12410 (R&S, SP, Security)
> www.MicronicsTraining.com <http://www.micronicstraining.com/>
> Sr. Technical Instructor
> YES! We take Cisco Learning Credits!
> Training And Remote Racks available

Blogs and organic groups at http://www.ccie.net
Received on Mon Sep 20 2010 - 18:18:34 ART