350-501 SPCOR Cisco CCNP Service Provider – MPLS Label Distribution Protocol part 2

3. Sharing Label Information

In this section, I’m going to explain you how the sharing of the label information is done inside the service for network. In most of the cases, like in the previous session, we have seen that MPLS is a technology which is going to enter as a normal IP packet and then it’s going to forward the information by seeing the label lookup instead of a normal IP lookup. But in order for that to happen, like the router should should be able to forward the packet based on the labels. There should be some label information has to be propagated within the service provider network. Now, there is something not happening automatically, so it doesn’t happen automatically all this label information.

If you want this process to happen, then it’s mandatory for you to configure some protocol called Label Distribution Protocol, which is responsible for sharing the label information. Now, in this section we are getting into some more in detail about how the MPLS responded. Devices share this label information. So the first thing, there are two protocols which are generally used for distribution of the protocols called TDP Tag Distribution Protocol and LDP. Now, before that, let me just get back to my label information. Just like it’s more similar to sharing the label information is more similar to your routing protocol. A routing protocol like Rip, Ehrposp, if you take any of these protocols. Let’s take an example. I got a 40 dot network.

Now the router one want to communicate with 40 dot network. Now, if I’m using any of the routing protocol, let’s say Rap or EHRP or OSP of how it is going to work. Now, by default, router D or Router Four is going to advertise 40 dot network to router next router that is enabled and this router advertises to another router and then this router advertise from the router. Now, router A is going to learn about the Fortune network and it’s going to install that particular route in the routing table with the next hop and the next stop information is P and the exit interface is this interface, right? In the same way, the LDP is also going to work now in LDP also it’s going to work in a more similar fashion. Now, here, let’s take an example.

I got a four door network here. Now, this four door network is automatically assigned with some label, something like local label. And this label information is advertised to the neighbor router, the neighbor LDB neighbor. And then now the router C is going to assign some local label to that and then it’s going to propagate the same information next router and then finally to the end router. Now, when this router want to reach Fortitude network instead of forwarding the packet based on the normal IP packet, it is not going to see the IP lookup, it’s going to see the label lookup. It’s going to see what is the local label assigned to that particular Fortitude network what is the outgoing label.

And based on that, it is going to communicate. Same thing happens here as well for the same 40 dot network what is the outgoing label and it’s going to forward based on the label. And the same thing, router C also will see what is the outgoing label for 40 dot network and then simply send out of the interface. So probably I’ll get into that more in detailed practical scenarios in my next video. But in this video, I’m just giving you some very basic information on the sharing of the label information. Now, in order to make this possible, you must configure an LDP protocol inside the service protocol network.

Now, there are majorly two different protocols which are certainly used. Tag distribution protocol which is a Cisco proprietary, very used, not much used today’s networks. And it works on TCP port numbers seven, one one. So initially Cisco introduced with some tag distribution protocol. Now even Cisco uses the default protocol as LDP, which is a standard distribution protocol, standard label distribution. And it is also the default label distribution protocol, default label protocol on Cisco devices and it works on UDP port number six process. Now, I can say that this is one of the one which is majorly used inside the service port network.

Apart from that, if you are using some advanced concept of MPLS, we call as MPLS traffic engineering, probably you’ll see in the service portal networks if you’re preparing for CCMP service portal or CCI service portal. So you’ll come across this topic where we have some traffic engineering and for traffic needing, it is going to add one more label. And that label binding is done by another protocol called RSVP. But again, inside the servicepone network majorly for forwarding the information we use any one of these protocols. But for traffic traffic engine tunnels it is going to add one more label and for that label it is going to use RSVP. So in this section we will be majorly focusing on LDP protocol, how to configure LDP protocol inside the service port networks.

4. MPLS LDP Configuration

Configuring LDP. Not a lot of theory. We are finally ready to start with our first lab of MPLS where we will see how to configure the label distribution protocol inside the service for network. Now, before we actually start configuring things, let’s try to understand what are the basic pre requirements we need to do before we actually confer LDP. Now the first requirement, the pre requirement here you can see that there must be a set enabled and we can verify with a command called Show ipset. Now Seth is a pre requirement. Why? Because like in Cisco devices we have something called Forwarding information based inside your data plane, built based on your routing table.

And based on this fib table, once you configure LTP, it is going to build something called Label Forwarding Information Base and any traffic entering the router it is forwarded based on by seeing verifying this table that is your Label Forwarding Information based table. So if this fib table is not built in case if your safe is not enabled, then there is no possibility of fib table is built and then there is no possibility of Lfib also will be built and there’s no way of sending the back end based on the labels. So Safe is a free requirement. We need to ensure, we need to check whether the safe is enabled or not. Now, we can verify with a command called IP Safe. You can see some kind of outputs here, you can see all the networks.

But in case if the safe is disabled, let me add one command no IP Safe. Now, if I disable the safe just I’m doing you’ll see this kind of message whenever you see this kind of message, you need to understand that Safe is not running on your routers. So whenever you see this kind of message you need to see you show Ipsaf. In most of the ibiz, in all the IBASE in the recent current iOS versions, safe is by default enabled. So we don’t need to generally add this command. But it’s always a good practice to just verify before you are going to start verify whether the safe is enabled or not. So this is what we did ypcef command if you see this message Safe is not running then we need to simply go and enable the Safe.

If you want to disable the safe we can use which is not recommended. So it’s an advanced fast forwarding mechanism which is by default enabled in the Cisco devices. Again, Safe is a pre requirement. Here the next thing we need to have an IGP running inside the service for network. Now IGP is mandatory because now just now we discussed that in the previous sessions whenever you advertise this particular network based on the routing table it is going to build to reach a specific destination, we need to have a routing table entry. So if I want to communicate inside my service phone network, from router one to router four. Now I need to go by a router two, router three and router four.

So there is a path and if you want to communicate the router one must have an entry for it or network so if there’s no entry for it or network then there is no way of your labeling work. So it’s something dependent. So you must have an IGP configured inside the service pointer which is going to build a routing table. And based on that routing table it is going to set is going to build something called Fib table forwarding information based and based on that Fib, once you configure label forwarding or label binding, then it is going to create something called Lfib. So here I didn’t configure this so I’m going to configure everything from the scratch here. Now on the router one I’m going to advertise the van interfaces and LAN interface.

And also I’m using loopback interface here, so I’m using one loop back interface. Also I’m advertising that loop back interface. It will be eleven one on the router one and 120 on the router 213 one on the router Three and 14 one on the router four. I think I have the diagram in my drawing here. Now here you can see the diagram. This is what I’m going to implement. So I’m going to configure the basic IGP inside the service for network I’m going to use OSPO and areas it can be any protocol you can use either OSPF rip EHRP ISS but considering the service for networks most commonly inside the service for MPLS code you’ll find either OSPF or ISS as the most widely used protocols because they are linked state and there is a feature called MPLS traffic engineering which is supported on OSPF and ISS.

So that was one of the reason why we most commonly use OSPF for ISS but you can use Rap or EHRP also. So in my scenario I’m going to use OSPF so let’s go to router one first. I’ll start with router one. On all the routers safe is enabled here so I’m going to say router OSPF one advertising the LAN interface ten all the eight subnet mask area zero and then advertising the Van interface and then advertising my eleven dot network. So you can advertise all the loopbacks whatever I have or you can just advertise 2. 5 area zero. So why am advertising the loopback? You’ll automatically come to know when I come discuss that label distribution protocol so the label distribution protocol neighbor ship establishes based on.

The loopback router ID so I’ll discuss more in detail about that concept but as of now you just advertise the land ban and the loop back interfaces as per the diagram. So on the router two I’m going to do the same thing advertising the LAN interface which is my 20 dot network 000-2508 so it will be 025-25-5255 N area zero and then advertising the one interface it is two and one. So 20 dot network, two dot network and one dot network and then the loop back into place. So I’m going to advertise all the loopbacks you cannot as I have four loopbacks pre configured here. So it will be twelve, zero, zero, it’s 24 subnet mask. Done. So I’m advertising my LAN interface, one interfaces and then finally the loop back interfaces.

So on the router three I’m going to do the same thing network three, dot network zero, dot 255-25-5255, area zero and then one face network three and two on the router three and then the loop back into 324 00:25 and then area zero. So just I finished configuring both the LAN interfaces and land interface and the load back interfaces. So if you want before you leave, if you want to verify, you can always verify with this command show IP OSPF interface brief. So both the one interfaces loadback interface and then the LAN interface. Same thing I’m going to do on the router four as well. Router OSP of one network, 40 dot network eight subnet mass and then three dot network and then advertising the loop back interface.

So it will be fourteen, zero, zero, zero, twenty five and then area zero. Done. So I just configured IGP that is OSPF inside the service for network. So I want to verify the neighborship first. So I’ll go router one. If I give show IP OSPF neighbors, I should see two neighbors. Similarly, I need to check on the router three as well. Show IP neighbor on the router three. Also I have two neighbors. If I go and verify my routing table on the router one, I should be able to see all the routes coming into my routing table. 40 coming from router 420 coming from router two and then 30 coming from router three. And these three are the loop back interfaces of router router two, three, four.

Okay, so similar way on the router tool so I can see all the routes coming. So it’s a basic basic IGP we need to configure and verify. Similar way if I use Show Ipsif, you can see Safe is going to populate the routing information based table that is rib table and then it’s going to create some FRB table for each and every network. Like in our case, I can verify with one network, let’s say Show IPSF 41 network and then the submarine mass slash H. I can see the route is learned via one or one or two and the exit interface is s zero by zero. That is the next stop.

5. MPLS LDP Configuration – Continued

Okay? So next thing we need to do is we need to configure LDP. So the pre requirements, the first thing safe must be enabled. We have verified that set is enabled and then we also configure IGP routing. So making sure that all the routers should be able to learn all the routes which are required. The first step we need to configure in LDP is so we just need to go to this configuration mode. We need to give this command MPLS label protocol LDP. Now this command is optional. And the reason it is optional because even though if you don’t configure this command, still by default LDP will be the default protocol which is used by Cisco devices.

So, which means this command I can see is optional even though if you don’t configure it takes LDP. But if you want to use other than LDP, let’s say if you don’t use TDP, you need to just write instead of LDP we need to write TDP. Okay? So the first command will decide which protocol you want to use for adding the labels. The next command. There is something called MPLS LDP router ID. And Router ID must be loop bag zero. Now this command is also optional, but it is more recommended command. Now the concept of router ID is more similar to OSPF. If you talk about OSP of Route ID or BGP router ID, it’s going to be the same.

So the first option always it will go is a manual route ID, right? In case if there is no manual route ready configured, then it is going to take the highest IP of the low back interfaces, right? Now in case if there is no loop back, let’s say then it is going to take the highest IP of the active physical interface, right? Now in case of LDP, also these conditions applies the same it’s going to be the same conditions what we have learned in the OSPF. But there’s one major condition here is the Router ID must be reachable, the route ID must be reachable. So the meaning of that is, let’s take an example. On the router one, we have taken the route ready as a loop back zero, but the same way on the router two, also we have taken the route ready as a loop back zero.

Now this router ID of the loop back router one must be able to reach the router ID of the router two, then only they will form LDP neighborship, then only they will exchange the labels. Then only your LDP is going to work. So this is one condition which is extra you will find in case of LDP routers. So we need to be very careful when we are selecting the router IDs because normally what happens if I don’t use this command? Now this command is also optional. If I don’t use this command by default, there is a possibility that it can take router ID any router. Let’s take an example. I’m going to ignore this command. I’m not going to write this command.

Now maybe you have some loopback zero which is let’s say on the router one is 110 zero one and also you have some loopback ten which is one seven to 16 one one. So if I don’t use manual route ready, in that case it is going to take the highest loop back interface. So in my scenario, 172 is higher than eleven. So which means now this will become the route ready. And if this loopback is not advertised inside your ICP, then you have reachable tissues and they will not form the neighbor ship and they will not exchange the labels. So there are two possible solutions in this kind of scenarios. Either you can advertise this interface inside your IGP, that is one solution, or you can just make a route ready loopback zero which is already advertised inside your IGP.

I prefer the first option, I always go with the route ready loopback zero. And because most of the scenarios we advertise the loop back zero IPS inside your IGP even we use in BGP as well where we use, where we form the neighbor shape with loop back interfaces. So there’s one extra thing we need to remember. So there’s a reason if you remember in our scenarios, like if you just get back to our router, in every router I have a low back interface one. And this low back interface is already advertised inside my IGP. So I did that already. You can see, even you can verify with this command, show IP OSP of interface brief. And in all my scenarios I’m going to use this loop bag zero as my router ID.

Then what are the commands we need to configure? So now the first command is optional, not compulsory. So you can skip this command because by default anyway, it is going to use LDP. And it’s always recommended to go with the next command because the router ID must be manually configured and that router ID must be advertised inside your IGP. Okay, so it’s mandatory. And then what are the commands we need to enable for label forwarding? Now we need to configure one single command called MPLS IP. Now we need to configure this command on each and every interface which is facing towards the service port network. In my scenario I’m going to assume that router one is going to connect to a customer, which is router five.

In the future classes we’ll see that and the router four, maybe it is connecting to some other customer router six. Now this customer is going to take the MP service from the service portal and router One, router Two, router Three, router Four, or acting as my service order networks. Okay, so now in this scenario what happens? The information enters into a normal IP packet and I want to ensure that when the router one sends inside the service for network, it has to be sent in the form of the labels. Now, which means we need to configure each and every interface which is facing towards the service for network must be enabled with MPLS IP command. Now the interfaces which I take here, these interfaces must be enabled with a command called MPLS IP.

Now, the customer facing interfaces, like in my scenario, this interface, it will be a normal IP packet. So we don’t need MPLS IP command on that interface. So customer facing interfaces will be a normal IP packet. So no need of MLS IP command. And the service portal facing interfaces must be forwarded in the form of a label. So we need to enable that command called MPLS IP. So if we don’t enable MPLS IP, in that case they will not form LDP Neighbor Shift and they will not exchange the label information and your label switch path will not work. Okay? So we are ready to go and configure these commands on our command line and then finally we’ll verify with Show MPs LDB enables, just like we verify the OSPF membership and then we’ll verify with Show MPLS interfaces.

It will show you what are the interfaces enabled with MPLS IP command. So let’s get started here. So the diagram, you can see the same diagram here. I already configured, IGP inside the service poor network. The first step, and the step is pre configured, so we don’t need to do anything safe, is by default enabled. So the second step, we are going to configure LDP inside the service poor network where we have a router one connecting to router two, router three, router four, and we don’t have any link connecting between one to four. Okay, if you want you can use that link. I’m not using it just for easy understanding. So let’s go to command line. So on the router one, the first command I’m going to use here is MPLS.

And apart from this commands, like whatever I have listed here, I’m going to use one more command here MPLS label range. And on the router one I’m going to use the range between 100 to 199 on the router two, I’m going to use the label range 200 to 299 and 300 to 399 on the router of three and 400 to 499 on the router four. Now, there is no specific significance for this assigning the labels here, but because normally zero to 15 are desert labels and the label range can be any number in between from 16 to any number, the router can assign any label to any network, each and every network it has to learn and it is just locally significant. So we don’t really need to define any range here.

But when it is going to assign some random labels, it becomes a little bit difficult for us to understand. So for easy understanding, I’m going to assign the label ranges of 100 and 189 so that I can understand. Okay, this label is assigned by Router One. So this label is assigned by Auto Two, Router Three, Router Four. But it’s really not required for you to configure in the production networks or in your lab scenarios. But in this first lab, I’m going to do that. So the first thing I’m going to say MBLs label range on the Router One, I’m going to say 100 to 199. And then Embellish LDP router ID must be loopback zero because I advertise the loopback zero in my IGP and it is reachable.

And then embellish label protocol LDP. So if I just use question mark here, you can see LDP LDP, the default protocol. So all the three commands are optional. But the last command, the middle command, is recommended command, but the remaining two commands are anyway not required. And then in my scenario, I need to enable which interface has MPLS IP here. Now I need to enable this S One by zero interface which is facing towards auto Two must be enabled with MPLS IP. Done. So if you want to verify Show MLS interfaces, you can see S One by zero is the interface is enabled with LDP. That’s it. So let’s go to router two again.

Let’s do the same thing. MPLS label range 200 to 299 for the Router Two mPLUS labels protocol LDP, MPLS LDP router loopbag zero. And here I’m going to assign mPLUS IP on two interfaces because I got s one by zero is connecting to router one and S one by one is connecting to router three. So we can see once I configure router two, interface s one by zero. Router one and router two. You can see the LDP neighborship messages comes up just like OSPF neighborship messages. Same thing I’m going to do here, 300 to 399399, and then the MPLS LDP are outright loop back zero. I’m not going to use mPLUS label protocol LDP because any vardp is default. So I simply say interface s one by zero, which is connecting to the router. Two MPLS IP.

And then interface s one by one. And then I’m going to say MGS IP done. The same thing I need to do on the Router Four as well. MPLS label range 400 to 499. MPLS LDP Router ID Loop bag zero. And then I just got only one interface, which is connecting inside the service for network. That is your hero one. So now I just configured MPLS LDP on these four routers. That verification time. So now, for verification, what I can do is I can just go to Router Two and I can give a command called Show MPLS LDP Neighbors. When I get show MLS LDP neighbors. I should see two neighbors. As for my scenario, router Two should form neighborship with Router One and Router Three. Now, similar way, if I go to Router Three, I should see two neighbors.

So that’s the first command we can verify. And the next command I’m going to use is Show MPLS interfaces. So let’s go to router two. Now I’m going to say show MPLS LDP neighbors. Now I can see, I’m able to see two neighbors. The first neighbor is route row 1110 zero one. And the next neighbor is 31. You can see this is a route Identifier value that is nothing but the route ready. And it’s mandatory that this Identifier value of the P and my local Identifier that is a local route ready. And the remote route ready must be reachable. Probably this is one of the troubleshooting you might need to do in case if your neighborship is not coming up, you need to check whether that outright is reachable or not.

A similar way, I’m going to verify on LDP neighborhood on the router three as well. I got router two as my neighbor and the router 14 as many birth. That is the router four. Next thing, there’s one more useful command you can use. You can always use show MPLS interfaces. This command gives some basic idea like what are the interfaces need to be enabled with MLS IP. And in my scenario, router three must have MPLS IP command on S one by zero and S one by one. So this is going to confirm that whether that interface is enabled with MPLS IP or not. So if this interface is not enabled with MLS IP command, then probably you don’t see that interface here and then you’ll also see what protocol it is going to use on that interface.

So if you see there is TDP running instead of LDP, then that might be the reason the neighborship will not come up because the protocol mismatch. One side if you confirm LDP, the other side if you confirm TDP, then they will not form the neighborhood. So there is something more in detail I’ll discuss in a separate section called MPLS Troubleshooting. But it’s important for you to understand that, okay, how to configure the basic LDP configurations and how to verify whether the LTP is configured correctly or not. So probably in my next section I’m going to give some more in detail about how the label forwarding is actually working in this scenario. So probably in any scenario it is going to work in the same manner.