Solved: Multicast does not work between differents VLAN

Martin Arsenault · ‎08-10-2011

Hello,

We try to activate multicast here and it does not seems to work properly.

There is our configuration :

Core : 4506 with L3 routing

Distribution : 3750 L2 only

Access: 2950 L2 only

On the 4506 we have those commands configured :

ip multicast routing

ip pim rp-address x.x.x.x where x.x.x.x=4506 IP address

on each VLAN, we have the command ip pim sparse-dense-mode activated.

We make our test with VLC Media player.

We are able to multicast on the same VLAN...but it does not seems to works between different vlan.

On the same VLAN we see the Networks Stream (SAP) annoucement appears in VLC (on the receiver PC), but on different VLAN, we don't see it. I've also try to force receiving by entering, on the receiver PC, the multicast address entered on VLC in the transmitter PC but that don't work.

For VLC, I've put 10 for TTL to be sure that all VLAN will be reached.

We have no access list configured on the two differents VLAN that we make our tests.

Someone can help?

Thanks,

Martin

Jon Marshall · ‎08-12-2011

Martin

Have you enabled pim on the vlan 1 interface that is the RP ie.

int vlan 1

ip pim sparse-dense-mode

Jon

View solution in original post

Jon Marshall · ‎08-10-2011

Martin

What multicast group number are you using ?

Also as a quick test to make sure it is not something else what happens if you try dense mode rather than sparse mode, does it work then ?

Jon

Edison Ortiz · ‎08-10-2011

Martin,

Can you provide the show ip mroute output from the 4506 switch?

Thanks

Martin Arsenault · ‎08-10-2011

Hi all,

Thanks for your fast reply. It's very appreciated.

I've made the test in Dense-mode and it's working. So why in Sparse-Dense-Mode that does not work ?

The transmitter PC is on VLAN12 and the receiver is on VLAN48

Here is the capture in Dense-mode for sh ip mroute 226.6.6.6

COREJ130#sh ip mroute 226.6.6.6
IP Multicast Routing Table
Flags: D - Dense, S - Sparse, B - Bidir Group, s - SSM Group, C - Connected,
       L - Local, P - Pruned, R - RP-bit set, F - Register flag,
       T - SPT-bit set, J - Join SPT, M - MSDP created entry,
       X - Proxy Join Timer Running, A - Candidate for MSDP Advertisement,
       U - URD, I - Received Source Specific Host Report, Z - Multicast Tunnel
       Y - Joined MDT-data group, y - Sending to MDT-data group
Outgoing interface flags: H - Hardware switched, A - Assert winner
Timers: Uptime/Expires
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 226.6.6.6), 00:36:44/00:02:51, RP x.x.x.x, flags: SJC
Incoming interface: Null, RPF nbr 0.0.0.0
Outgoing interface list:
Vlan48, Forward/Dense, 00:00:09/00:00:00
Vlan12, Forward/Dense, 00:28:52/00:00:00

(y.y.y.y, 226.6.6.6), 00:00:16/00:02:56, flags: T
Incoming interface: Vlan12, RPF nbr 0.0.0.0
Outgoing interface list:
Vlan48, Forward/Dense, 00:00:09/00:00:00, H

Where

x.x.x.x = 4506 Ip address

y.y.y.y = transmitter PC ip address

There is the capture in Sparse-Dense-mode for sh ip mroute 226.6.6.6

COREJ130#sh ip mroute 226.6.6.6
IP Multicast Routing Table
Flags: D - Dense, S - Sparse, B - Bidir Group, s - SSM Group, C - Connected,
       L - Local, P - Pruned, R - RP-bit set, F - Register flag,
       T - SPT-bit set, J - Join SPT, M - MSDP created entry,
       X - Proxy Join Timer Running, A - Candidate for MSDP Advertisement,
       U - URD, I - Received Source Specific Host Report, Z - Multicast Tunnel
       Y - Joined MDT-data group, y - Sending to MDT-data group
Outgoing interface flags: H - Hardware switched, A - Assert winner
Timers: Uptime/Expires
Interface state: Interface, Next-Hop or VCD, State/Mode

(*, 226.6.6.6), 00:39:56/00:02:40, RP x.x.x.x, flags: SP
Incoming interface: Null, RPF nbr 0.0.0.0
Outgoing interface list: Null

(y.y.y.y, 226.6.6.6), 00:03:28/00:02:55, flags: PT
Incoming interface: Vlan12, RPF nbr 0.0.0.0
Outgoing interface list: Null

Where

x.x.x.x = 4506 Ip address

y.y.y.y = transmitter PC ip address

Thanks for your help.

Martin

Edison Ortiz · ‎08-10-2011

Do you have multiple Core switches?

If so, which switch is running the RP?

It could be the receivers were joining the mcast tree to a different RP.

On the 2nd output, it indicates the S,G is pruned due to SPT-bit set.

With dense, it will flood every single switch and that's why it is working.

Martin Arsenault · ‎08-10-2011

Hi Edison,

We have two core switch running HSRP : Hostname are CoreJ130 & CoreD003.

We also use PVST mode to feed each distribution switch (3750). Each 3750 have one link to CoreD003 and another link to CoreJ130.

The captures sent earlier are from the CoreJ130 and the IP listed x.x.x.x are the CoreJ130 IP address. On the CoreD003 switch, we have those commands configured :

ip multicast-routing

ip pim rp-address x.x.x.x where x.x.x.x is the CoreJ130 address.

We also have ip pim sparse-dense-mode on all VLAN interface on the CoreD003 switch.

So it appears that we have only one RP. The two core switch has the same RP address configured pointing to the CoreJ130 core switch.

Jon Marshall · ‎08-10-2011

Martin

Can you confirm that both core switches have -

1) the rp address configured (sounds like it does)

2) ip pim sparse-dense-mode under the L3 vlan interfaces on both core switches.

Jon

Martin Arsenault · ‎08-10-2011

Hi Jon,

I confirm :

Both Core switch have ip pim rp-address x.x.x.x --> x.x.x.x = CoreJ130 ip address.
Both VLAN interfaces (on CoreJ130 and CoreD003) have ip pim sparse-dense-mode configured.

Martin

Edison Ortiz · ‎08-10-2011

On the HSRP setup, both Vlans are set to the same physical switch as the primary HSRP?

I'm assuming the RP is the loopback from the Core switch, can you ping the loopback from the other Core switch?

You didn't mentioned any routing between them.

Martin Arsenault · ‎08-10-2011

HSRP is configured in this way :

x.x.x.2 CoreJ130

x.x.x.3 CodeD003

x.x.x.1 HRSP address

VLAN are on the same physical switch / primary HSRP (x.x.x.2).

There is a trunk configured between the two core switch. There is no routing between core switch.

There is no loopback interface created on core switch. Does it requires a loopback interface? Does the RP address must be a loopback interface ?

Jon Marshall · ‎08-10-2011

Can you just confirm that you can ping the RP address from the core switch that does not have the actual RP address interface on it ?

Jon

Martin Arsenault · ‎08-10-2011

Yes, I confirm that I can ping RP address from the CoreD003.

Jon Marshall · ‎08-10-2011

Martin

Sorry, should have been more specific, can you ping the RP address on the RP switch using the source IP address of the vlan 12 interface ?

Jon

Martin Arsenault · ‎08-11-2011

Jon,

From the PC on the VLAN12 I can ping the CoreJ130 (x.x.x.2) but I cannot from the PC on VLAN48 because there is an ACL that limit access to CoreSwitch. I made the test from my PC that have access to CoreSwitch.

Do I have to create a loopback interface that can be reached from all PC in our network and define it as the RP ?

Is the RP must be visible from all network ?

Jon Marshall · ‎08-11-2011

Martin

No the PCs do not need to be able to route to the RP just the actual routers and there is no requirement that the RP be a loopback address but it usually is because if you use an interface and that interface fails you have lost the RP whereas if you use a loopback that interface is always up as long as the switch.

Can you post -

1) vlan interface configuration from both core switches

2) if there are acls applied to interfaces can you post those too

3) "sh ip route" from both core switches

4) specify the IP address of the source and receiver

Jon