LabMinutes# SP0030 – Cisco MPLS Multicast VPN (Part 1)

therefore welcomed lemons common olam video streaks on MPLS you can find ended list of MPLS video on a website by click on the link above and sign up for our newsletters to receive the latest video revises in this video we’re going to be configuring MPLS multicast VPN the first gonna start configuring the world-wide multicast routing for the ambulance structure and we exclusively going to be using specific multicast for that they will look at the two most important elements of MPLS multicast VPNs and they are default MDT or data MDT and NDT stands for organized thrown distribution tree and you will see how they are used differently and why would you want to use the data MDT in addition to the default and then we’re gonna be enabling per vrf multicast routing I’m gonna try both a dense and sparse mode for that for a physical laboratory topology we have a drought arrggh r1 through our 8+ 1 switch switch one with our 2 3 4& 5 connected to cross point-to-point see pertains while the other routers and substitution are connected to cross layer tube as shown in this diagram okay for our bed three topology we have a pretty much a standard MPLS VPN setup with the Alcor MPLS networking within this orange circle and r1 r2 and r4 being the PE routers and now three our five being the P routers and we already have the ibgp hearing going for our MP BGP between PE routers and for our customer sites “were having” three places they are being represented by a router or 6 or 7 and permutation 1 and they are running BGP for the pece routing protocol and all these devices have to leap back 10 through 12 praise and then communicate it across the MPLS VPN so I’ll go in this lab is to provide Malik as rich clevernes among these three sites so the MPLS multicast VPN is one of those concepts that’s a little convoluted and it’s always a good meaning to have a complete understanding before we get into the actual configuration other words you could be easily devastated by the concept and the amount of configuration that’s compelled so I’m gonna do first is to kind of walk you through at the high-level overview of what’s really going to happen as part of the MPLS VPN to support the Maori assigns so let me bring up something that I can kind of draw on and let’s go ahead and step through that so the first thing that you would need to do is to enable your MPLS core to support Madi shoots and that’s basically your underlying infrastructure so making our sip situation has an example that means you have to enable all the links that we’ve seen the connects out the P and P routers together with pimps and that is just the basic multicast routing to be enable and the next you need to create or define a multicast under the customer vrf and this including defining your default MDT Mountie cache address as well as address reserve for the data MDT and mostly what the default MDT is is the first multicast distribution tree that’s being joined by all of the PT routers and that’s how they communicate whether it’s a dominate or data traffic chiefly so it’s kind of a overlay on top of the world multicast routing that’s only being participated by the PE routers that have the multicast routing in a wonder the vrf okay the data MDT on the other hand is something that will be constructed later once the data traffic is start flowing through across the network and that’s to provide an optimal path for route high-throughput data traffic’s which we will explain in a second how that will be constructed later but first thing first is for the called appear adders to join a default MDT and that can usually happens in two ways one is using the share tree as part of the sparse mode and that would have to be bi-directional share tree so you just go through the whole rendezvous quality descriptions just how you build a podcast network with the sparse mode but it has to be bi-directional since any of these PE routers are most likely be both the source and listeners of the traffic a second approach you can also use for this which is a more recommended or commonly used comings to use the source specific multicast which is what we’re going to be doing in this lab but for the source for multicast to work their listener needs to know the IPS of the resources from a out-of-band method and then formerly that the IP is known the listener can send a John request directly to the source it doesn’t have to go through a share tree and for this particular scenario with the MPLS VPN natty Kass the default MDT IPs and the source Pei Pease mapping will be communicated as part of the MP BGP and you participate last-minute that for that to be supported for that a newer iOS version you need to enable the address lineage ipv4 MDT and formerly the PE routers kind of discover each other they will immediately build a sauce located tree and they will act as both being a source and a listener ok so just imagine that you have chipped an overlay with default MDT that each of these PE routers will join and then what once that’s been erected your next have to enable multicast locally on each of these sites where there’s being a thick-witted most division smaller even source-specific multicast and then a source of the multicast group can pretty much start channelling to any other sites over the default MDT once today the issue of trafficking can excess the threshold that you configure for the data MDT the sauce P will a situate a new multicast address from a data M DT pool and then start advertising that to all of the neighbours and that circular going to contain the mapping between the of the source PE routers and the multicast group in that action and the other PE routers that has a neighbourhood listener can send a source specific multicast meet to the source PE and mostly fabricate a direct beginning based tree for that data traffic of the particular multicast group ok for the PE router does not have the local listeners is basically going to cache that planning in case in the future if we received the John request from the regional invention ok so let’s say in this case r6 is the cause and it starts transporting traffic initially will go across the default MDT but once the threshold is outstripped let’s say this which one is the only listener for those traffic then it will basically build a data MDT instantly exactly to a router one while you router to completely stay away from that data MDT because it does not have local listener and that highway the amount of data traffic doesn’t get prepared given to all of the PE unnecessarily okay so that would be over the data MDT okay so that’s pretty much what’s happening behind the situation so now we can do is to go ahead with our configuration exercises so our lab scenario here is to configure it all of our PE routers to provide multicast connectivity between Mount Ebal MPLS VPN areas yeah the test number one is to deal with world malady statu routing and first we need to configure multicast route on router one two three four and five utilizing the sparse mode and then we need to configure all of our PE routers to support source-specific multicast but only for the following non-default MDT groups and we want our default MDT to multicast group to be within those IP arrays two three nine zero zero zero reduce 24 for our data MDT we wanted to be within the two three 9.25 5/16 let’s work our space from left to right starting with our first PE router router r1 it’s to enable multicast routing you do IP multicast routing then we need to configure a PM sparse mode on MPLS interface only fastened 0 zero as well as the loop back because this source interface for our bgp discussions and that’s specific requirements if you’re using the resources specific multicast okay since that’s going to become the source off to traffic as the mapping is sent out to all the PE machines so for our loopback 0 this to LP kim sparse and then fast 0 0 IP pim sparse then we have to come up with an access index to define now SSM multicast group it’s just going to call this MVT and our first assortment of Monte Casa address is two three nine zero zero zero and that is last-place 24 and for our second multicast group range for the data MDT is two three nine two five five zero zero/ 16 and then we restrain that to a dominate IP pin SSM as you can see the question mark you have two alternatives here for us is to default so if you were to use the default SSM multicast radical which is 2 3 2/ 8 for your MDT’s then you don’t really have to go through these practise merely creating access listings but since we are using the non default SSM mounted shed stray we need to kind of override the default exerting our assortment dominate and then type the access listing to that so MDT okay let me try to copy that into a notepad since we’re going to have to configure the same thing on several devices pastes and I also need the authority for SSM just going to positioned that together ok and and too Neath the IP multicast routing merely kind of kind of bill a speedy template here so that should be it for our one next moving on to the router r3 r3 needs three interfaces to be enable actually I’m just going to go ahead and enable the move back as well or the magic ass so is moving forward and replica this and friendly back to zero character in sparse procedure and then we also have three additional boundaries fast 0 0 0 0 0 and then 0 1 0 although it needs to pay man able transcript and then paste looks like I misspelled the sparse let’s try that again there you go so I’ll be out on three Nexus our router out for same thing required three interfaces enable plus the loopback so it would be serial zero zero zero this place one zero and then when it’s to call on zero actually an scribe we didn’t really need to define the MDT because it’s not a PE router that will participate in SSM let me actually remove that okay but for our off more our for its going to participate in the default MDT so we need to have that enable and then for our 5 are 5 PE routers all we need is the multicast route enable on all the boundaries and so our 5 paste and then on r2 r2 is al PE routers so needed the MDT range command let me modify the interface real quick okay they’re just replica and on our to paste okay so so far we just enabled my basic multicast routing with PIM sparse mode time do a immediate substantiate IP PIM neighbor and experience our two is being a neighbor for r4 and r5 let’ s check our three really got to make sure same thing do you present IP PIM neighbor and we should be seeing our one four and five being our three neighbors okay and that should end out research number one

You May Also Like