All things related to multicast

[William Caban <>]

Well, here it is.

UPR: University of Puerto Rico

Setup Scenario
===========

We have a Cisco GSR serving all our campuses and this is the one that is 
doing the peering with Ampath for I2 services. On unicast we only 
advertise /24s of those campuses that are part of the University and not 
the /16 of our main class (136.145/16) since we have (for historic 
reasons) private entities being which Internet services are serve by 
this class (we are working on fixing that). On multicast we just 
advertise the whole /16 since these external private entities are not 
multicast enable anyway.

(AMPATH/I2)-----[GSR at UPR]----(UPR I2 institution)
                                   \-----[7600 a] -------- (other 
smaller UPR institutions)
                                                 \-------------[7600 b] 
---- (Internet - Local Carrier)
                                                                       
\----- (private entities)


So here I'll be referring to the GSR, UPR I2 institutions and the "7600 a".


Actual information
=============

From the GSR I have an MBGP peering with AMPATH in which I advertise 
the 136.145/16 plus the Arecibo's Observatory Networks, the output for 
this part:

GSR12406#sh ip mbgp neighbors 198.32.252.209 advertised-routes
BGP table version is 988206, local router ID is 136.145.215.50
Status codes: s suppressed, d damped, h history, * valid, > best, i - 
internal,
             r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

  Network          Next Hop            Metric LocPrf Weight Path
*> 136.145.0.0      0.0.0.0                            32768 i
*> 192.65.176.0     136.145.215.18           0         32768 ?
*> 192.231.93.0     136.145.215.18           0         32768 ?
*> 192.231.95.0     136.145.215.18           0         32768 ?

From the GSR I also have an msdp peering with AMPATH, the UPR I2 
campuses that want Multicast and to the 7600:

GSR12406#sh ip msdp summary
MSDP Peer Status Summary
Peer Address     AS    State    Uptime/  Reset SA    Peer Name
                               Downtime Count Count
198.32.252.33    20080 Up       3w0d     2     2724  gsr.ampath.net
136.145.215.14   65003 Down     00:00:05 0     0     
rrp2hpcf.hpcf.upr.edu  !They were down on this snapshot
136.145.215.6    65002 Up       6d00h    1     0     rcm2hpcf.hpcf.upr.edu
136.145.215.49   5786  Up       4d22h    0     0     gw7600.hpcf.upr.edu

Now, I seen on the beacon (http://beacon2.ncsa.uiuc.edu/) that the two 
beacons, on two different campuses are behaving quite different:

ciar-mcast.rcm.*upr*.*edu (*136.145.128.147) >> connected on a campus, 
mostly have multicast connectivity
wisepiont.hpcf.upr.edu (136.145.54.40) >> connected to the 7600, have no 
multicast connectivity

The IGMP on the 7600 look like this:

7609#sh ip igmp groups 233.4.200.18
IGMP Connected Group Membership
Group Address    Interface                Uptime    Expires   Last Reporter
233.4.200.18     Vlan2                    4d19h     00:02:01  136.145.54.40
7609#

7609#sh ip mroute 233.4.200.18 136.145.54.40 count
IP Multicast Statistics
12 routes using 8100 bytes of memory
8 groups, 0.50 average sources per group
Forwarding Counts: Pkt Count/Pkts per second/Avg Pkt Size/Kilobits per 
second
Other counts: Total/RPF failed/Other drops(OIF-null, rate-limit etc)

Group: 233.4.200.18, Source count: 2, Packets forwarded: 5158036, 
Packets received: 5158036
 Source: 136.145.54.40/32, Forwarding: 5158036/13/55/6, Other: 5158036/0/0
7609#


Now on the GSR I get this:

GSR12406#sh ip mroute 233.4.200.18 136.145.54.40 count
IP Multicast Statistics
161 routes using 77456 bytes of memory
9 groups, 16.88 average sources per group
Forwarding Counts: Pkt Count/Pkts per second/Avg Pkt Size/Kilobits per 
second
Other counts: Total/RPF failed/Other drops(OIF-null, rate-limit etc)

Group: 233.4.200.18, Source count: 51, Group pkt count: 9603068
 Source: 136.145.54.40/32, Forwarding: 0/-12/0/0, Other: 5158454/0/5158454


What does "-12" means!? How can I forward packets in neg? The RPF for 
this network on these routes look like this:

GSR12406#sh ip rpf 136.145.54.40
RPF information for wisepoint.hpcf.upr.edu (136.145.54.40)
 RPF interface: GigabitEthernet1/2
 RPF neighbor: ? (136.145.215.54)
 RPF route/mask: 136.145.54.0/24
 RPF type: unicast (ospf 5786)
 RPF recursion count: 0
 Doing distance-preferred lookups across tables
GSR12406#

RPF information for wisepoint.hpcf.upr.edu (136.145.54.40)
 RPF interface: Vlan2
 RPF neighbor: ? (136.145.54.40) - directly connected
 RPF route/mask: 136.145.54.0/24
 RPF type: unicast (connected)
 RPF recursion count: 0
 Doing distance-preferred lookups across tables
7609#

At the beacon page wisepoint.hpcf.upr.edu (136.145.54.40) is clasified 
as a Blind Beacon. The other machine, ciar-mcast.rcm.*upr*.*edu* 
(136.145.128.147) from time to time disappear from beacons like 
*.indiana.edu.

Where in the configuration do I start looking for possible explanations?


Thanks in advance,

--William Caban

William Caban wrote:

> I need help debugging our multicast (University of Puerto Rico). I 
> remember I did the multicast configuration back in 2002 and basically 
> haven't touch it since then. Lately we have been having various 
> strange behavior where in a short period of time (an hour) people see 
> us from some places and then it stops. I've try my best and I think it 
> should be working (but clearly is not). Will this be a forum for this 
> type of questions to post more detailed information? or, can anyone 
> point me to the right forum?
>
>
> Thanks in advance,
>
> William Caban