Thursday, December 4, 2008

How default mdt and data mdt works?

Prior reading to the this kindly have a look on Basic Of MVPN.

Default mdt is used by multicast tunnel in MVPN to carry customer multicast traffic. It is unique per vpn. By defining default mdt under every vrf a multicast tunnel interface establish with every PE. So in short we can it is a full mesh process by default. Cisco uses GRE tunnel for multicast interfaces tunnel and MTU is of 1514 bytes. You cannot edit and delete the tunnel interface until and unless delete the default mdt. Check the interface tunnel with show interface command you will be weird to see one thing the destination address will be the default mdt group address with source address loopback of the PE router. From this it is very much clear any default mdt which is being created for any vpn should be reachable in service provider with the help of loopbacks of PEs. Why loopbacks because they are used as next hop in MP-iBGP. It means default mdt will flood the multicast traffic on each and every multicast tunnel whether the multicast is required by the PE or not. So it means *,G entry is created and unnecessary flood of multicast traffic in network which can cause lot of unnecessary problems.
To overcome the problems originated by default mdt will be sorted out by using data mdt. By usinf data mdt we can set the rate limit of originating PE. If will send the message encapsulate in default mdt to every multicast tunnel by saying whether you required multicast stream or not. If PE responds with positive nature a S,G entry is created in the router. So the next time when the stream will come instead of sending the stream to each and every PE it will forward the multicast stream only to those which joins data mdt. It means it will check the S,G entry.
In short we can say default mdt always create *,G entry and data mdt always create S,G entry.

shivlu jain

People who read this post also read :

No comments: