Multicast routing

[Cisconew]

Hi,

Below is my task to do. I have configured as per below network topology. Unicast routing is working from receiver to source and vice versa. But multicast routing is not working. I need to know the reasons for this and also the task question also.Please help me.

Task:

Only BGP is running on the routers. Each router is in its own ASN. Only the LAN networks (where servers are connected) and the loopback0 addresses are advertised via BGP using network statements. BGP synchronization is disabled. Transit networks are not advertised. There are no default networks. PIM sparse mode is configured on all interfaces and the RP is set to 3.3.3.3(Router 3) on all routers.
Router 3 is hitting 100% CPU utilization and dropping packets because of the high CPU. What could be causing this?

[Harold Ritter · Accepted solution]

Hi @Cisconew ,

Here's probably what is causing the high CPU on the RP (R3).

1. Multicast traffic coming from Source1 is received on R1

2. R1 encapsulates the multicast packets and send them to R3 (PIM register message with source 192.168.12.1) until it receives a PIM register stop message.

3. R3 receives the encapsulated packets from R1 and tries to send a PIM register stop message to 192.168.12.1, as this is the PIM register message source address.

4. R3 doesn't have a route to 192.168.12.1, as transit interfaces are not advertised in BGP and can't send the PIM register stop message to R1

5. it keeps receiving the PIM register messages and need to decapsulate lots of traffic, which is causing the high CPU.

Regards, 

[Cisconew]

Hi,

Thank you for your clarification.

R1 is sending Multicast packets from its

G 0/1

interface to R3's Loopback address(3.3.3.3) which is advertised in BGP.

But R3 sends stop message to R1's  physical interface not loopback interface(1.1.1.1).

So, in this case, Is it possible  Receiver  can receive Multicast stream directly from source?

[Harold Ritter]

Hi @Cisconew ,

In spite of the situation I described, the receiver should still be able to receive the multicast traffic from the sender.

I think the issue is that your multicast receiver is misconfigured. It should have any multicast configuration, except for the

ip igmp join-group 229.7.7.7

as it is acting as a host not a multicast router.

Regards,

[Cisconew]

What is the Solution for the above issue.

Should I advertise transit networks in BGP? OR any other solution?

[Harold Ritter · Accepted solution]

Hi @Cisconew ,

There is two ways you can fix this issue.

1. use the command

ip pim register-source Loopback0

on R1. You also need to configure

ip pim sparse-mode on lo0

2. Inject the transit subnets in BGP.

I prefer option 1.

Regards,

[Cisconew]

Hi 

I have used option 1 

ip pim register-source Loopback0

on R1. Also, I have configured 

ip pim sparse-mode on R1's lo0.

Then I simulated the ping from source(pinging to 229.7.7.7.But it didn't receive replies.Also, I have checked

sh ip mroute

in R1,R3,R4.

Source#ping 229.7.7.7 repeat 10000
Type escape sequence to abort.
Sending 10000, 100-byte ICMP Echos to 229.7.7.7, timeout is 2 seconds:
...............................................

R1#sh ip mroute | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 229.7.7.7), 00:00:17/stopped, RP 3.3.3.3, flags: SP
Incoming interface: GigabitEthernet0/2, RPF nbr 192.168.14.4
Outgoing interface list: Null

(10.10.10.10, 229.7.7.7), 00:00:17/00:02:42, flags: PT
Incoming interface: GigabitEthernet0/0, RPF nbr 10.10.10.10
Outgoing interface list: Null

(*, 224.0.1.40), 02:45:27/00:02:35, RP 3.3.3.3, flags: SJPL
Incoming interface: GigabitEthernet0/2, RPF nbr 192.168.14.4
Outgoing interface list: Null

R3#sh ip mroute | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 224.0.1.40), 02:50:39/00:03:02, RP 3.3.3.3, flags: SJCL
Incoming interface: Null, RPF nbr 0.0.0.0
Outgoing interface list:
GigabitEthernet0/3, Forward/Sparse, 02:49:29/00:02:58
GigabitEthernet0/2, Forward/Sparse, 02:50:39/00:03:02

R4#sh ip mroute | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 229.7.7.7), 02:51:08/00:02:58, RP 3.3.3.3, flags: SP
Incoming interface: GigabitEthernet0/3, RPF nbr 192.168.34.3

R4#sh ip mroute | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 229.7.7.7), 02:51:08/00:02:58, RP 3.3.3.3, flags: SP
Incoming interface: GigabitEthernet0/3, RPF nbr 192.168.34.3
Outgoing interface list: Null

(*, 224.0.1.40), 02:51:08/00:02:58, RP 3.3.3.3, flags: SJPL
Incoming interface: GigabitEthernet0/3, RPF nbr 192.168.34.3
Outgoing interface list: Null

(*, 224.0.1.40), 02:51:08/00:02:58, RP 3.3.3.3, flags: SJPL
Incoming interface: GigabitEthernet0/3, RPF nbr 192.168.34.3
Outgoing interface list: Null

[Cisconew]

I have wiped out all my old configurations and reconfigured everything.Now multicast is working.

First, I didn't issue

ip pim register-source Loopback0

command in R1.so it continuously got register message in

sh ip route

output . After configuring this command, Now register message disappeared and now it is working as expected.

Thank you all for guiding me.

This the output

R1#sh ip mrou | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 228.7.7.7), 00:01:09/stopped, RP 3.3.3.3, flags: SPF
Incoming interface: GigabitEthernet0/1, RPF nbr 192.168.12.2
Outgoing interface list: Null

(10.1.6.6, 228.7.7.7), 00:01:09/00:01:50, flags: FT
Incoming interface: GigabitEthernet0/0, RPF nbr 0.0.0.0, Registering
Outgoing interface list:
GigabitEthernet0/2, Forward/Sparse, 00:00:53/00:02:36

(*, 224.0.1.40), 00:12:42/00:02:21, RP 3.3.3.3, flags: SJCL
Incoming interface: GigabitEthernet0/1, RPF nbr 192.168.12.2
Outgoing interface list:
Loopback0, Forward/Sparse, 00:12:41/00:02:21

After issuing "ip pim register-source Loopback0"" command registering stopped.

R1#sh ip mrou | b Int
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 228.7.7.7), 00:04:17/stopped, RP 3.3.3.3, flags: SPF
Incoming interface: GigabitEthernet0/1, RPF nbr 192.168.12.2
Outgoing interface list: Null

(10.1.6.6, 228.7.7.7), 00:04:17/00:03:23, flags: FT
Incoming interface: GigabitEthernet0/0, RPF nbr 0.0.0.0
Outgoing interface list:
GigabitEthernet0/2, Forward/Sparse, 00:04:01/00:03:24

(*, 224.0.1.40), 00:15:50/00:02:12, RP 3.3.3.3, flags: SJCL
Incoming interface: GigabitEthernet0/1, RPF nbr 192.168.12.2
Outgoing interface list:
Loopback0, Forward/Sparse, 00:15:49/00:02:12




Source#ping 228.7.7.7 repeat 10000
Type escape sequence to abort.
Sending 10000, 100-byte ICMP Echos to 228.7.7.7, timeout is 2 seconds:
...................................................................
Reply to request 67 from 10.4.5.5, 31 ms
Reply to request 68 from 10.4.5.5, 9 ms
Reply to request 68 from 10.4.5.5, 27 ms
Reply to request 69 from 10.4.5.5, 6 ms
Reply to request 69 from 10.4.5.5, 19 ms

Reply to request 70 from 10.4.5.5, 6 ms
Reply to request 71 from 10.4.5.5, 12 m

[MHM Cisco World]

10.4.5.5 

this IP of receiver ?

[Cisconew]

yes.

That is IP address of Receiver.

[MHM Cisco World]

Great' glad your lab now working.

Now why I suspect of bgp path' the answer is urpf.

I was thinking and still that path go and back through same interface may make urpf drop multicast.

It need some deep dive.

But later

Have a happy summer 

Thanks 

MHM

[MHM Cisco World]

if I am right 
change the RP from R3 to R2 make RP address 2.2.2.2 
and check again 
let see if I am right.
NOTE:- you must change RP in all routers

[MHM Cisco World]

5 min it take from me to do lab 
I notice you two times
change rp to be 2.2.2.2
lab and success no issue. 

[MHM Cisco World]

If I not reply sorry I watch movie with my family.

Write your Q and i will check later 

Thanks 

MHM

[Harold Ritter]

Hi @MHM Cisco World ,

You should be able the same lab with R3 as the RP.

Regards,

[MHM Cisco World]

@Harold Ritter  as always was right
I config LO in IOU3 (3.3.3.3) and make it

rp-address

and as you can see the multicast is success.