No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Problem of PIM-DM multicast routing trough NE40

Publication Date:  2012-07-27 Views:  40 Downloads:  1
Issue Description
There is a Head End Station broadcasting a package of TV channels and IP/MPLS Core network of the customer, which transmits video to the final IPTV subscriber. There is a need to transmit 12 channels into the MPLS network (multicast group: 239.255.0.1 � 239.255.0.12) irrespective of whether there are subscribers or not. For this purpose the PIM-DM protocol and constant presence imitation of IGMP subscribers on Ge 1/0/2 interface are used. Brief details of configuration of the NE40 router dividing Head End Station and MPLS network are given below:
…..
#
 multicast routing-enable   
…..
#
interface Vlanif902
 undo shutdown
 ip address 10.1.0.1 255.255.255.252 
 igmp enable
 pim dm
…..
interface Ethernet2/0/14.102
 vlan-type dot1q vid 102
 description MulticastLink
 undo shutdown
 ip address 10.0.0.2 255.255.255.0 
 pim dm
……
interface GigabitEthernet1/0/2
 portswitch
 undo negotiation auto
 undo shutdown
 port trunk allow-pass vlan 901 to 903
 igmp host-join bridge 239.255.0.1 vlan 902
 igmp host-join bridge 239.255.0.2 vlan 902
 igmp host-join bridge 239.255.0.3 vlan 902
 igmp host-join bridge 239.255.0.4 vlan 902
 igmp host-join bridge 239.255.0.5 vlan 902
 igmp host-join bridge 239.255.0.6 vlan 902
 igmp host-join bridge 239.255.0.8 vlan 902
 igmp host-join bridge 239.255.0.9 vlan 902
 igmp host-join bridge 239.255.0.10 vlan 902
 igmp host-join bridge 239.255.0.11 vlan 902
 igmp host-join bridge 239.255.0.7 vlan 902
#
Alarm Information
No packages are being transmitted into the MPLS network except the multicast group 239.255.0.1. As a result IPTV subscribers  can watch only the first TV channel. 
After analyzing the multicast routing table we can see the following:
[Irkutsk_NE40_ATE27-Ethernet2/0/14.102]disp multicast routing-table 
Multicast Routing Table
Total 12 entries
(10.0.0.1, 239.255.0.1)
    Uptime: 00:00:07, Timeout in 293 sec
    Upstream interface: Ethernet2/0/14.102(10.0.0.2)
    Downstream interface list:
      Vlanif902(10.1.0.1), Protocol 0x3: IGMP NBR
(10.0.0.1, 239.255.0.2)
    Uptime: 00:00:07, Timeout in 293 sec
    Upstream interface: Ethernet2/0/14.102(10.0.0.2)
    Downstream interface list:
      Vlanif902(10.1.0.1), Protocol 0x2: NBR
…….
…….
 (10.0.0.1, 239.255.0.12)                  
    Uptime: 00:00:42, Timeout in 271 sec  
    Upstream interface: Ethernet2/0/14.102(10.0.0.2)
    Downstream interface list:            
      Vlanif902(10.1.0.1), Protocol 0x2: NBR
                                          
Matched 12 entries                        
After analyzing the multicast forwarding-table  we can see the following:
[Irkutsk_NE40_ATE27-Ethernet2/0/14.102]disp multicast forwarding-table 
Multicast Forwarding Cache Table
Total 12 entries
00001. (10.0.0.1, 239.255.0.9), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 129688 pkts(16602496 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00002. (10.0.0.1, 239.255.0.5), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114485 pkts(14656512 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00003. (10.0.0.1, 239.255.0.7), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114507 pkts(14658112 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00004. (10.0.0.1, 239.255.0.12), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:          
       01: Vlanif902
     Matched 114534 pkts(14661568 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00005. (10.0.0.1, 239.255.0.11), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114583 pkts(14669056 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00006. (10.0.0.1, 239.255.0.6), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 133729 pkts(17119744 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00007. (10.0.0.1, 239.255.0.3), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114667 pkts(14678592 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00008. (10.0.0.1, 239.255.0.8), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114679 pkts(14680128 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00009. (10.0.0.1, 239.255.0.2), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 114715 pkts(14684736 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00010. (10.0.0.1, 239.255.0.10), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 124531 pkts(15941184 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
00011. (10.0.0.1, 239.255.0.1), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 124538 pkts(15942080 bytes), Wrong If 0 pkts
     Forwarded 124480 pkts(15934656 bytes)
                                          
00012. (10.0.0.1, 239.255.0.4), iif Ethernet2/0/14.102, 1 oifs
     List of outgoing interface:
       01: Vlanif902
     Matched 124563 pkts(15945280 bytes), Wrong If 0 pkts
     Forwarded 0 pkts(0 bytes)
      
Handling Process
After the disable the PIM DM protocol on Gigabitethernet 1/0/2 and Ethernet 2/0/14.102 interfaces and enable once again  - this problem was solved. The above described problem happened because after the command: igmp host-join bridge 239.255.0.1 vlan 902  the pim dm protocol was enabled on interfaces and all other host-join were written later.
Root Cause
As we can see from the routing and forwarding tables, NE40 doesn't transmit multicast packages from the Ethernet 2/0/14.102 interface into the Gigabitethernet 1/0/2, if they do not belong to the 239.255.0.1 group. It happens because the PIM-DM protocol believes that there are IGMP subscribers on the Gigabitethernet 1/0/2 interface only for the 239.255.0.1 group (See: Protocol 0x3: IGMP NBR), and аs for the other groups, there is only the neighbour multicast router (See: Protocol 0x2: NBR). However, while configuring we gave  commands to NE40 to imitate constant presence of IGMP subscribers on the interface for all the 12 groups. For what reason PIM-DM protocol perceived only the first cammand:
igmp host-join bridge 239.255.0.1 vlan 902 and ignored all the rest analogous commands.
Suggestions
After a large number of experiments it was ascertained that the commands, imitating the constant presence of IGMP subscribers on interfaces must be given before the command "pim dm" which enables multicast routing protocol on interfaces. If PIM-DM is already enabled and the command igmp host-join will appear for the group X.X.X.X. later, this command will be ignored by the PIM-DM protocol and it will reject multicast traffic until the IGMP request is received for the X.X.X.X group from the MPLS network.

END