From: Scott Morris (swm@emanon.com)
Date: Mon Feb 02 2004 - 15:41:10 GMT-3
Making the hub the mapping agent is perfectly acceptable as well, because
you have two different groups for a reason. The first group 224.0.1.39 is
the dense mode group that the RP will announce itself on. This will run
into issues previously discussed about OIL and the fact that if you hear it
from one interface it won't go back out the same interface.
That's where the second group 224.0.1.40 comes into play. This is the
mapping agents' discovery notices. The concept of an OIL here still works,
but not with regards to other groups. So if you hear something from a frame
spoke on 224.0.1.39, it has nothing to do with sending an announcement out
that same interface to 224.0.1.40.
Now, point to point subinterfaces certainly make things much easier, but
that would also take all the fun out of the lab design! :)
Placing the mapping agent BEHIND the hub will work as well, but is not
necessary as part of the solution.
Scott Morris, CCIE4 (R&S/ISP-Dial/Security/Service Provider) #4713, CISSP,
JNCIS, et al.
IPExpert CCIE Program Manager
IPExpert Sr. Technical Instructor
swm@emanon.com/smorris@ipexpert.net
http://www.ipexpert.net
-----Original Message-----
From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf Of
Brian McGahan
Sent: Monday, February 02, 2004 1:23 PM
To: alsontra@hotmail.com; ccielab@groupstudy.com
Subject: RE: Multicast Question
Alsontra,
It's not an RPF error, but instead a fundamental rule of multicast
forwarding. Think of it like split-horizon for multicast.
When running auto-rp, the groups 224.0.1.39 and 224.0.1.40 must be
joined to listen for candidate rp and mapping agent advertisements. The
common workaround for this issue is to run either sparse-dense mode so that
these groups are dense, or to enable autorp listener. The problem with
either of these solutions is that there is not an explicit sparse join, so
the OIL is associated with the interface, not the IP address when running
NBMA mode.
This implies that the mapping agent and candidate RPs must be placed
in a manner that all points of the network can receive a feed from them. In
the case of hub and spoke this implies that they are placed behind the hub.
Designwise the best workaround is just to run point-to-point IP
subnets. In the lab this could be accomplished by running GRE tunnels.
HTH,
Brian McGahan, CCIE #8593
bmcgahan@internetworkexpert.com
Internetwork Expert, Inc.
http://www.InternetworkExpert.com
Toll Free: 877-224-8987
Direct: 708-362-1418 (Outside the US and Canada)
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On Behalf
> Of alsontra@hotmail.com
> Sent: Monday, February 02, 2004 1:17 PM
> To: ccielab@groupstudy.com
> Subject: Re: Multicast Question
>
> If your using this type configuration with PIM-SM, the RP and mapping
> agent must be configured on the HUB router, correct? How does PIM in
> dense mode deal with this config? You have to build tunnels on the
> HUB to satisfy RPF, right????
>
>
> Alsontra
>
>
>
> ----- Original Message -----
> From: "Brian McGahan" <bmcgahan@internetworkexpert.com>
> To: "'Bob Sinclair'" <bsin@cox.net>; "'Ashok Verma (ashoverm)'"
> <ashoverm@cisco.com>
> Cc: <ccielab@groupstudy.com>
> Sent: Monday, February 02, 2004 8:53 AM
> Subject: RE: Multicast Question
>
>
> > Technically a multicast feed can be sent out the same interface it
> > is received on. This is in the case of PIM sparse running in NBMA mode.
> >
> > With NBMA mode, the OIL is associated by peer IP address, not the
> > outgoing interface. Therefore in this case when R1 receives a feed
> > from
> R2,
> > the incoming interface is seen as R2's IP address, not the interface
> > connected to the Frame Relay cloud, and the outgoing interface to R3
> > is
> R3's
> > IP address, not the interface connected to the Frame Relay cloud.
> >
> >
> > HTH,
> >
> > Brian McGahan, CCIE #8593
> > bmcgahan@internetworkexpert.com
> >
> > Internetwork Expert, Inc.
> > http://www.InternetworkExpert.com
> > Toll Free: 877-224-8987
> > Direct: 708-362-1418 (Outside the US and Canada)
> >
> >
> > > -----Original Message-----
> > > From: nobody@groupstudy.com [mailto:nobody@groupstudy.com] On
> > > Behalf
> Of
> > > Bob Sinclair
> > > Sent: Monday, February 02, 2004 6:59 AM
> > > To: Ashok Verma (ashoverm)
> > > Cc: ccielab@groupstudy.com
> > > Subject: Re: Multicast Question
> > >
> > > Ashok,
> > >
> > > By design, PIM sparse and dense modes will not send traffic out
> > > the incoming interface. This is part of the PIM method for
> > > creating loop free
> trees.
> > > More formally, the incoming interface will not show up on the
> > > outgoing interface list of the S,G entry.
> > >
> > > In order to get the multicast stream through R1 to R2, you will
> > > need
> to
> > > have
> > > it arrive one one R1 interface (or subinterface) and leave on another.
> > >
> > > HTH,
> > >
> > > Bob Sinclair
> > > CCIE #10427, CISSP, MCSE
> > > www.netmasterclass.net
> > >
> > >
> > >
> > > ----- Original Message -----
> > > From: "Ashok Verma (ashoverm)" <ashoverm@cisco.com>
> > > To: <ccielab@groupstudy.com>
> > > Sent: Monday, February 02, 2004 5:00 AM
> > > Subject: Multicast Question
> > >
> > >
> > > > Hi All ,
> > > >
> > > > I have a very simple setup of Muticast,
> > > >
> > > > R1
> > > > / \
> > > > / \
> > > > R2 R3
> > > >
> > > > FR cloud is used between these routers and R1 is Hub, other
> > > > router
> are
> > > > spoke.
> > > >
> > > > If I enable multicast in on all three router and join a
> > > > multicast group(224.8.8.2)on R2, I am able to ping that group
> > > > from R1 but not
> from
> > > > R3, I had enabled the Ip Pim dense mode on all the interfaces as
> > > > required.
> > > >
> > > > So question is Do I need to do some more configuraiton on R1 so
> > > > that
> R3
> > > > should able to ping the multicast group.
> > > >
> > > > Thanks in advance.
> > > >
> > > > Ashok Kumar Verma
> > > > CCIP,CCDP,CCNP
> > > > Network Consulting Engineer
> > > > Customer Advocacy Advanced Service Dep.
> > > > Service Provider AS Div.2
> > > > Cisco Systems, K.K.Japan.
> > > >
> > > > Tel: +81-3-5324-4583
> > > > e-mail: ashoverm@cisco.com
> > > >
> > > >
> ______________________________________________________________________
> _
> > > > Please help support GroupStudy by purchasing your study
> > > > materials
> from:
> > > > http://shop.groupstudy.com
> > > >
> > > > Subscription information may be found at:
> > > > http://www.groupstudy.com/list/CCIELab.html
> > >
> > >
> ______________________________________________________________________
> _
> > > Please help support GroupStudy by purchasing your study materials
> from:
> > > http://shop.groupstudy.com
> > >
> > > Subscription information may be found at:
> > > http://www.groupstudy.com/list/CCIELab.html
> >
> > ____________________________________________________________________
> > ___ Please help support GroupStudy by purchasing your study
> > materials from:
> > http://shop.groupstudy.com
> >
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
> ______________________________________________________________________
> _ Please help support GroupStudy by purchasing your study materials
> from:
> http://shop.groupstudy.com
>
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
This archive was generated by hypermail 2.1.4 : Fri Mar 05 2004 - 07:13:45 GMT-3