L2TPV3 and multicasting

Unanswered Question
Sep 30th, 2010
User Badges:

We have configured L2TPv3 between two routers R1 an d R2 on eth0/0. R1 and R2 are connected by L3 network.


Host1----> eth0/0 of R1----> ( L3 network ) -----------> R2-eth0/0 --------> Host2


Following configuration:
l2tp-class l2tp-defaults
retransmit initial retries 30
cookie-size 8
pseudowire-class ether-pw
encapsulation l2tpv3
protocol none
ip local interface Loopback0

Iinterface Ethernet 0/0
xconnect 10.0.3.201 123 encapsulation l2tpv3 manual pw-class ether-pw
l2tp id 222 111
l2tp cookie local 4 54321
l2tp cookie remote 4 12345
l2tp hello l2tp-defaults


And corresponding configuration on R2 also.


If we connect hosts to eth0/0 of R1 and eth0/0 of R2 with IP address in the same subnet, they can ping to each other.
That means frames arriving on eth0/0 of R1 are encapsulated by L2TPv3 protocol and then are decapsulated on Eth0/0 of R2.

But when host generates the multicast-traffic, then multicastframes or packets are not getting encapsulated by L2TPV3 protocol as those are
not received on the other end of the virtual tunnel.


Host1----> eth0/0 of R1----> ( L3 network ) -----------> R2-eth0/0 --------> Host2

Host1 can Ping host2  ( from 10.1.1.1 /24 to 10.1.1.2 /24 ).


We replaced the hosts by routers and enabled OSPF on the corresponding interfaces on the routers, OSPF nbr relationship was not coming up.
Debugging shows that it is sending the multicast hello packet but it is not receiving the corresponding hello packet from other router.

Multicasting frames are getting encapsulated.


Please share the experience.
Thanks in advance
Subodh

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Calin Chiorean Mon, 10/04/2010 - 00:28
User Badges:
  • Silver, 250 points or more

Hello Subodh,


I will simulate your config in my environment to see if I get the same behavior and share the results with you.


I just need one more detail. Between R1 and R2, where you have the L3 connection, you have some IGP running in which you advertise the Loopback interface?


Cheers,

Calin

Calin Chiorean Mon, 10/04/2010 - 00:49
User Badges:
  • Silver, 250 points or more

Hello!


I have tried your configuration and for me it's working. I have configured OSPF on the edge routers and it's working without any issue. Below you have parts of the config that are related to this troubleshooting.


My topology is R3(edge)F0/0 <-> Fa0/0 R1 S0/0<-> S0/0 R2 Fa0/0 <-> Fa0/0 R4(edge)


R3 Fa0/0 10.10.34.3 /24

R4 Fa0/0 10.10.34.4 /24


On S0/0 interfaces of R1 and R2 I have a point-to-point connection and Loopback (R1 - 10.0.3.201, R2 - 10.0.3.202) addresses of R1 and R2 are routed over L3 link to have reachability:


## R1 ##

l2tp-class l2tp-defaults
retransmit initial retries 30
cookie size 8
pseudowire-class ether-pw
encapsulation l2tpv3
protocol none
ip local interface Loopback0


interface FastEthernet0/0
no ip address
duplex auto
speed auto
xconnect 10.0.3.202 123 encapsulation l2tpv3 manual pw-class ether-pw
  l2tp id 111 222
  l2tp cookie local 4 54321
  l2tp cookie remote 4 12345
  l2tp hello l2tp-defaults


## R2 ##

l2tp-class l2tp-defaults
retransmit initial retries 30
cookie size 8

pseudowire-class ether-pw
encapsulation l2tpv3
protocol none
ip local interface Loopback0


interface FastEthernet0/0
no ip address
duplex auto
speed auto
xconnect 10.0.3.201 123 encapsulation l2tpv3 manual pw-class ether-pw
  l2tp id 222 111
  l2tp cookie local 4 12345
  l2tp cookie remote 4 54321
  l2tp hello l2tp-defaults



## R3 ##

R3#sh ip os nei


Neighbor ID     Pri   State           Dead Time   Address         Interface
10.10.34.4        1   FULL/DR         00:00:30    10.10.34.4      FastEthernet0/0


## R4 ##


R4#sh ip ospf nei


Neighbor ID     Pri   State           Dead Time   Address         Interface
10.10.34.3        1   FULL/BDR        00:00:33    10.10.34.3      FastEthernet0/0


Are you sure that you don't have anything else that deny multicast traffic or some minor mistake that you have missed there?


Cheers,

Calin

bapatsubodh Mon, 10/04/2010 - 06:21
User Badges:

Hi,


First of all thanks a lot that you tested the set-up for me, really appreciate it.


We are running OSPF between our backbone network. So the loopback of one router is reachable to another router loopback by OSPf proces.

As in your example we are running OSPF between R1 and R2 as back bone protocol.

And just a little change, we are not directly connecting the R3 and R4 to the R1 and R2, we have L2 switch in in between.

Second observation is OSPF is formed but hellos are not received by one router. So neighbor relationship is valid for default dead time  period. It all builds the neighbor from start.

Debugging shows that both routers send hello messages but only one receives and other don't. So naturally it delacres the neighbor as dead and starts the process all ove ragain.


I am not clear why this is taking place, my guess is that both routers send hello to the same Destination multicast addres. Still not clear on this point.


Thanks a lot again.

Please share your experience.

Subodh

Actions

This Discussion