cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2094
Views
0
Helpful
4
Replies

Multicast inter-vlan traffic intermittently not forwarded to receiver host

Hi,

I've configured our network for multicast routing but on one of our building network ( L3 with 3750 and access layer with 2950 switches), the multicast traffic is intermittently not forwarded between 2 vlans.

The senders are located in a vlan 192, the receiver is located on a vlan 249. Both vlans are routed on the same 3750.

The IGMP group registration seems to work fine. The multicast mroute table looks good but sometimes, the receiver does not see the multicast traffic arriving on the port. It's intermittent. Sometimes, it receive the multicast traffic sometimes not. ( the traffic concern video camera over IP)

Could you please help me to solve the issue ?

Network config is the following

Distribution layer - two  3750 with 2 vlans 192 and 249 (in HSRP)

Access layer - 2950 switches

4 Replies 4

Mohamed Sobair
Level 7
Level 7

Hi,

You will need to configure the Mapping Agent for the Multicast groups in order to recieve thier mapping information.

your config is OK, please configure the following on the 3750 and post the output od (sh ip mroute) (sh ip pim rp mapping) (sh ip agmp groups) after the changes:

ip pim send-rp-discovery Loopback1 scope 10 group-list 4

let us know if this solves your issue,

Mohamed

Hi Mohamed,

THanks for you reply.

The RP mapping agent is already configured on our Core switches ( C6509)

The network topology looks like this

Client vlan 249 and sender (ip camera) vlan 192 --> L2 C2950 --> L3 switches C3750-M249 (2 redundant C3750) --> Core Switch (Mapping agent) --> L3 Switch C3750-C67 (other building) --> C2950 --> Other Client vlan 67 and other cameras vlan 192

What happens is that clients in vlan 67 routed at 3750-C67 can well see cameras located on vlan 192 of the building M249.

But clients in vlan 249 get troubles to see the cameras in the vlan 192 routed on the same C3750-M249. Sometimes it works, sometimes disconnected/reconnected a lot.

Note that a client in vlan67 can see cameras located in vlan 192 (of C67 building) routed at C3750-C67

I already configured the C6509 as mapping agent for the multicast groups 239.255.24.0/24.

I don't understand why it works fine on building where vlan 67 and 193 and Not on the other.

Are you sure I need to configure the first 3750 as a mapping agent since I already have mapping agent configured on the COre switch ?

Configuration on the core switch:

ip pim send-rp-discovery Loopback1 scope 10

C6509#  sh ip pim rp mapping 239.255.24.6
PIM Group-to-RP Mappings
This system is an RP-mapping agent (Loopback1)

Group(s) 239.255.24.0/24
  RP 172.16.1.244 (?), v2v1
    Info source: 172.16.1.244 (?), elected via Auto-RP -->  This is well the L3 3750 switch of the M249
         Uptime: 3d19h, expires: 00:02:44
  RP 172.16.1.243 (?), v2v1
    Info source: 172.16.1.243 (?), via Auto-RP
         Uptime: 1w1d, expires: 00:02:52

DISTRI1_M249#sh ip pim rp
Group: 239.255.24.6, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.7, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.8, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.9, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.10, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.11, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.12, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.13, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24
Group: 239.255.24.14, RP: 172.16.1.244, v2, v1, next RP-reachable in 00:00:24

DISTRI2_M249#sh ip pim rp
Group: 239.255.24.6, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.7, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.8, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.9, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.10, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.11, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.12, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.13, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.14, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51

On DISTRI1_M249 :

DISTRI2_M249#sh ip pim rp
Group: 239.255.24.6, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.7, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.8, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.9, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.10, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.11, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.12, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.13, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51
Group: 239.255.24.14, RP: 172.16.1.244, v2, v1, uptime 3d19h, expires 00:02:51

I'll give you the mroute status later because the multicast client getting issue is not working. SO no mroute to display...

But I can confirm you that the outgoing and incoming interfaces of the mroute is correct

It looks like on DISTRI1_M249

(10.24.50.246, 239.255.24.6), 16:19:44/00:02:37, flags: T
  Incoming interface: Vlan249, RPF nbr 0.0.0.0
  Outgoing interface list:
    Vlan192, Forward/Sparse-Dense, 16:19:44/00:02:53

The client 10.24.50.246 is located in vlan 249 and the sender is well located in vlan 192

I cannot see any differences between a working situation and a non working one. The mroute table seems always correct.

Thanks a lot.

Regards

Mohamed Sobair
Level 7
Level 7

Hi Fredric,

If mapping Agent configured on the core, then no need to configure it on the 3750, However, you said the mroute table is Ok, I dont see its OK, there is multicast routing issue on the 3750.

The RPF interface shouldnt display an address of 0.0.0.0, there should be an RPF neighbor though,

Please post the Mroute table again from the 3750 and the core Switch as well as from the client router side.

Mohamed

semihuzun1
Level 1
Level 1

Hi,

Please refer to this document it seems that you experiencing some expire timer issue:

d. S,G expiry timer

Finally, Cisco has made a modification to the operation of the S, G expiry timer in IOS. There is now a CLI knob to allow the state for a S, G to stay alive for hours without any traffic being sent. This fix was in response to a customer request in a PIM-SM environment to maintain the state and not fall back to *, G forwarding. The command is "ip pim sparse sg-expiry-timer" and is documented in the command reference:

http://www.cisco.com/en/US/docs/ios/ipmulti/command/reference/imc_04.html#wp1018443

This approach should be considered a workaround until PIM-Bidir or PIM-SSM is deployed or the app is fixed.

http://www.cisco.com/en/US/prod/collateral/iosswrel/ps6537/ps6552/ps6592/whitepaper_c11-474791.html

Review Cisco Networking for a $25 gift card