From: David Knot (david_knot@xxxxxxxxx)
Date: Mon Nov 05 2001 - 12:41:04 GMT-3
I've built the same scenario. I've got it working by
putting mcast on all interfaces of all mcast routers.
Doyle vol 2 says that you should put mcast on all
interfaces of all mcast routers otherwise SPF
calculation can fail.
--- Sijbren Beukenkamp <s.beukenkamp@nl.ibm.com>
wrote:
> This does not help at all.
> Question for all: Try to rebuild the senario, as I
> have already tried
> several other senario from " Developing IP Multicast
> Network VI, author
> Beau Williamson, Cisco Press" and I am unable to get
> them to work
> properly.
>
> By the way: If you want to know everything about
> multicast. I can
> recommend the book.
>
> Looking forward
> Met vriendelijke groet/ Kind regards,
> Sijbren
> ++++++++++++++++++++++++++++++++++++++++++++++++++++
> Ing. Sijbren Beukenkamp IBM Global Services
> Internetworking Specialist Integrated Technology
> Services
>
> Networking & Connectivity
> Services
> http://www.ibm.com/services/nc/
> GSM +31 (0)6 53761703 The Netherlands
> Office: +31 (0)30 2850 666
>
>
>
>
> Brian Hescock <bhescock@cisco.com>
> Sent by: nobody@groupstudy.com
> 31-10-2001 14:26
> Please respond to Brian Hescock
>
>
> To: Sijbren
> Beukenkamp/Netherlands/IBM@IBMNL
> cc: ccielab@groupstudy.com
> Subject: Re: Multicast Sparse-Dense
> Mode
>
>
>
> Pinging from R1 means the packets will be
> process-switched. When pinging
> from R4 the packets on R1
> will be fast-switched. Configure "no ip
> mroute-cache" on the outgoing
> interface on R1 and then see
> if you can ping from R4. If not, put an ip igmp
> join-group on the
> outgoing interface on R1 and see
> if you can ping from R4 and get a response from R1.
>
> There have been several multicast fast-switching
> bugs where sometimes you
> don't even get an
> interface in the OIL when you have a join-group
> command on that interface.
> As a workaround, what
> will often work is removing join-group and remove
> pim, let pim timeout
> then put the commands back
> on. If that doesn't work, do the same thing again
> but once the commands
> are off the interface do a
> "shut", wait a few seconds, do a "no shut" and put
> the commands back on
> the interface. You should
> then see the outbound interface in the OIL. This
> isn't something you
> should have to do, just a
> workaround due to a multicast fast-switching bug.
>
> Brian
>
> Sijbren Beukenkamp wrote:
>
> > Consider the following.
> >
> > Three routers R2, R1 and R4. All interconnected
> using FR. R1 is in the
> > centre and will have to route Multicast between
> its serial interface to
> R2
> > and R4. R2 and R4 have a tokenring interfaces at
> which SD-mode is
> > configured. The same is for all the serial
> interfaces.
> >
> >
>
LAN--------R2-------------FR---------------R1----------FR---------------R4-----
---LAN
> >
> > IP MULTICAST-ROUTING IP
> MULTICAST-ROUTING IP
> MULTICAST-ROUTING
> > PIM-S-D
> PIM-S-D PIM-S-D
> > igmp-join 224.1.1.1
> > 224.1.1.1 announced by RP
> > igmp-join 224.2.2.2
> >
> > R2 is RP and mapping agent ONLY for 224.1.1.1. So
> basicly 224.1.1.1 is
> > using Sparse-mode and 224.2.2.2 is using Dense !
> > PIM Sparse Dense is configured on all Frame-relay
> interfaces and the
> > tokenring interfaces.
> > On the tokenring interface of R2 are the two igmp
> joins configured.
> >
> > R1 and R4 are showing the RP (sh ip pim rp
> mapping)
> > R1 can ping 224.1.1.1 and 224.2.2.2
> > R4 can only ping 224.2.2.2
> >
> >
> > What's wrong with this picture !
> >
> > Met vriendelijke groet/ Kind regards,
> > Sijbren
> >
> ++++++++++++++++++++++++++++++++++++++++++++++++++++
> > Ing. Sijbren Beukenkamp IBM Global Services
> > Internetworking Specialist Integrated
> Technology Services
> >
> Networking & Connectivity
> > Services
> >
> http://www.ibm.com/services/nc/
> > GSM +31 (0)6 53761703 The Netherlands
> > Office: +31 (0)30 2850 666
>
This archive was generated by hypermail 2.1.4 : Fri Jun 21 2002 - 06:45:04 GMT-3