Re: blocking mapping agent

Hi Piotr

From the doccd:

"IP multicast boundaries filter data and control plane traffic including
IGMP, PIM, and Auto-RP messages. PIM Register messages are sent using
unicast and will *not *be filtered."

As your tests showed, routing protocol traffic sent to multicast addresses
is also not filtered, which is the desired behaviour.

Regards
Roy

On 28 August 2011 05:21, Piotr Malarski <piotr.malarski.99_at_gmail.com> wrote:

> Hello guys,
> I was testing mapping agent blocking in following topology
>
> R4----R5(s1/1)------(s1/1)R6
>
> OSPF is routing protocol
> R5 is mapping agent.
> Task says to configure R5 so R6 does not receive RP announcement
> On R5 I did:
> # access-list 1 deny any
> # int s1/1
> # ip multicast boundary 1
>
> I tested this knowing this is not a right config. Deny any in access
> list should block ALL multicast, at least what I thought. In this case
> OSPF hellos should be blocked and OSPF should break. Interestingly,
> OSPF works fine and I can see in debug that R5 and R6 send and receive
> 224.0.0.5 successfully. R6 does not receive mapping agent messages as
> expected.
>
> It looks like "ip multicast boundary" does not stop 224.0.0.5.
> Is there an implicit permit in command "ip multicast boundary", Is it
> for all 224.0.0.0/24 ?
>
> Thanks,
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>

-- 
Regards
Roy
Blogs and organic groups at http://www.ccie.net