[Xorp-users] How to work with a bridge (br0)? (corrected version)

Pavlin Radoslavov pavlin@icir.org
Wed, 23 Nov 2005 13:12:55 -0800


> > 1. Configure the XORP router to be the RP for some particular group
> >    range. The simplest way to do it is to use the "static-rps"
> >    statement.
> > 2. Start receivers directly connected to each of the
> >    interfaces/segments: eth0, eth1, eth2.
> > 3. Start a sender on each of the interfaces/segments (eth0, eth1,
> >    eth2) and verify that the receivers receive the data.
> 
> What address should I enter as RP? Which of my own? I am confused.

It shouldn't matter which of your RP addresses you will enter.
Note that the whole purpose of the above test is to verify that you
send and receive across the XORP router. However, from your email
below it looks like you can perform a similar test using
senders/receivers spread across your domain so you can skip the test
I suggested.

> The current (best) result is following:
> 
> 1) Sender is inside the local network
> 2) This PC sends SAP announcements and data
> 3) SAP announcements can be seen university-wide
> 4) Data cannot be seen
> 5) Data sent to the SAP address (sap.mcast.net) can be seen (works
> flawlessly!)

Before doing anything, first double-check that the TTL of the data
is large enough, because the default multicast TTL is 1.

Then check the MFC entries ("show pim mfc") whether there is a
matching entry for your PC/sender and the group address for the
data that is missing. You could double-check by running the UNIX
"cat /proc/net/ip_mr_cache" command to see what exactly is installed
in the kernel.

If there is an entry, check that the iif and the oifs are
correct. If they are correct, then run tcpdump on one of the
expected outgoing interface and see whether the data is coming out
on that interface.
If there is no MFC entry, then use the "show pim join" command to
check what is the status of the corresponding (S,G) entry for this
directly connected source.

Further debugging of the problem will depend on the above results.

Pavlin

> 
> So I am quite confused. I think my xorp configuration is quite OK by
> now. Sending works (with that address) and receiving works with all
> addresses. Sending from the outer network (so next to xorp's external
> interface) does not work with normal addresses - just the same as from
> inside.
> 
> Should I ask my university? My university's provider? Is my
> configuration incorrect? Am I using the wrong addresses? I tried
> several, including 233.x.y.z where x and y are derived from my
> university's AS number.