OSPF Areas

3. Multiple Areas - Design Rules

So the next thing I'll discuss is the rules. Like in the previous section, we discussed why we generally go with multiple areas. As I said, to minimise the size of the database, to minimise the number of adjustments Once we decided to go with multiple areas, we needed to follow some rules, like the designing rules. So like I said, if you remember, in the previous basic topics, we have used only one area. So, in most cases, we can stick to one area. It's like if you have 30 to 50 routers, that's what Cisco recommend number. Let's say you have 30 to 50 routers and each router has, let's say, three to four interfaces, roughly around 100 networks.

So you can go with one area. So for any number less than 30 to 50 routers, we can go with a single area where you can just make all the routers in one single area, let's say 30 plus routers. And that area number can be any number. You can use either area zero or just area 10. So when you're using one single area, you can use any number. Okay, let's say if my design goes more than 100 routers, like I said, if you have some 30 plus routers, I can go with any area number, like I can go with area ten or area 20 or area 30. But when your design goes beyond that point, let's say I've got 100 plus routers and each router has three to four subnets, so it goes around three to 400 subnets. Now, we need to go with multiple areas again because if you don't go with multiple areas again, there are some problems which we discussed that will increase the secretization. So those problems again come into the picture. So we want our network to perform more and better, so we need to go with multiple areas. So the first rule is that whenever you decide to go with multiple areas, like two, three, four, any number, So I decided to add some routers. Let's say these are the routers, and I decided to go with another area, like 30 routers here, 30 routers here, 30, 40, like that, 30 to 50, that's our recommended number.

So let's say I'm adding 30+ routers here in another area. Now, whenever you go with two areas, then the second area should be area zero, which means I can go with area ten. Any number you can use, like when we give some numbers for the areas and the numbers can be any number. You can probably use any number normally, but when we use two areas, then you cannot say I'll use area ten and area twenty. Now this is wrong; it's not going to work. So basically, whenever you go with two areas, one area should be zero. That means you can design area ten and area zero, or you can use any number here, but the second number should be zero compulsory. So there is a rule. So if I say I'll go with ten and 20, that is not going to work. Area zero is mandatory whenever we go with two or more than two areas. So, if you must have one area called areazero, and you have two or more, you generally refer to this area zero as a backbone area, a backbonearea where all the other areas are connected. Again, I'm coming to the backbone design next. The area zero is referred to as the "backbone area." That is technically referred to as "bag Bonaria." Okay? So that is your second rule. So the first rule is that when you use one area, you can use any number.

And the second rule is that when we use two or more than two areas, the area zero should be present compulsively. Now the third rule is all the non-backbone areas. Now non-backbone means any other area other than zero. We call them as nonbibble. Like, let's say I'm creating area 30, area 40, and area 50, like that and some routers here. So basically all these areas are called "non-backbone areas." So when I say non-backbone, it means other than area zero. So the third rule says that all non-backbone areas must connect to area zero, which means any other area should be using area zero as a transit area. Now, what does that exactly mean? Well, let's say I decided to design one topology. Let's say I got some routers. This is how they are connected. Let's say the connectivitywise and I got some routers like this. So this is just a rough topology. How are my routers connected?

Let's say So there are plenty of numbers. So I decided to go with multiple areas. So I decided that, okay, this part on the left side, I'm going to assign them an area of ten. And then I decided to add another area. Let's say I decided to use this part as area zero. That is good. And I decided that this part may have 30 to 40 routers here decided as area 20. Let's say any number you can use apart from zero. Zero is something you must have and apart from that, the remaining numbers can be any number. And now I decided that this particular part, this leftover part, I want them to make another area to group them in another area. So let's say I'm going with area 30 here. Now this area is probably not going to work. And the reason is when you're designing. So every nonbagbone area must connect to areazero, which means it has to connect to any router that belongs to area zero. So basically, what it means is that every nonbag's own area when you're designing it should be like this, something like this. So one router like this So something like this So area zero should be the backbone area or the transit area. That means when any traffic is coming from area 30, let's say 30, or let's say 50, here, it should go to the router.

This is the router which belongs to zero, and from there it will go to the next area, maybe area 60, let's say. So it's not like it should go via area zero, it should go via any one router in area zero. So if you see this design here, now area 30 is not connected to area zero. So it's actually connecting to Area 20. And this area is probably not going to work. This means whatever the route advertisements say, they will not go to other areas by default. So again, there is another solution called virtual link. We'll probably talk about that in the advanced topics probably. But normally, as per the default design, this area 30 is not going to work because this is not as per the rules. So when you're designing, let's say this is your one design. Here you can see that, basically, I'm just making this all-head office router and the branch offices in area zero. And then I'm going to assign this router as a border router. And all the routers in this branch office, let's say this is in one of the cities, connect to small towns here, probably in Area 10, this in Area 20, and this in Area 30. Now this design works because when this router, routerone, let's say, wants to communicate with router two, So it goes to this bottle router and this router belongs to area zero also. So from there, it will go directly here if you have a connection, or it will go from this site, and then it will communicate something like this. So when you're designing this, you have to keep this in mind.

As a result, every non-backbone area must connect to area zero, or any route of area zero should connect. So that is like the compulsory rule. So in this design, if I want to add more routers or in a new area, let's say I need to have something like this. So let's say this is how they are connected. I've got some routers. I can assign these to one area, let's say area 50. So likewise, I can design one more area, area 70, some routers here. Okay, so I'm not drawing the routers, but you can probably have multiple routers in that. And of course, this area is not going to work. So basically, in this example, this area is not going to work because area 30 is not connected to area zero. So if you want to connect again, you need to have one router connected to any router. Now you can make area 30 like this. Now, this will work if you design it this way. If you don't do that, then it's not going to work. So again, this is the rule you need to keep in mind. And the next rule is that you need to have at least one border router.

Now, in general, the router that connects Here In this example, you can see this router is called an area borderrouter because this router is a member of area one. So one interface in area one and another interface in area zero means it is going to receive the advertisements for both the areas and also maintain the database of both areas. Generally, if you verify the IPO database, So this router will participate in both the areas' algorithms. So you must have at least one borderrouter, and you may have more than one. And this border router is responsible for exchanging routes between both areas because there should be one common router which participates in both the area advertisements and the area bottle out. So how to identify the router which is a member of both areas? Like in this example here. So this is your bottle cap. Because if you observe here, this router has one interface in area 50 and another interface in area 51. So this is your bottle outer and here is your bottle outer. And here, maybe this is your bottle router.

Okay? So, typically the bottle outer is a router which connects to two or more than two areas. Maybe it also connects to one more area here. So it depends. Like you can see, there are four areas it can connect or five areas, depending on your design. But generally, when we say border router, it means it is connecting to more than one area. So you must have at least one bottle outer compulsory. So at least one interface in both areas is going to connect both areas. In general, this is one more rule you need to keep in mind. So the final rule is that the interfaces facing each other must be in the same area. So, which means, let's say I have these three interfaces, three routers. Now these opposite interfaces, these two, must be in the same area. So as per the design, let's say I'm assigning this in area ten, the opposite interface also should be in area ten. If I'm measuring this interface in area20 or zero, let's say the oppositeinterface also should be in area zero. But let's say if I design means advertising, if I advertise this interface in area ten and the opposite interface is in area zero or area 20 or area 30,other than the same area, the neighbourhood will not be established. So when they exchange the hollow messages initially, the neighbour ship will establish based on the hollow messaging. If you remember the twoways, they will also exchange some additional information and that information has to match. If there is an area mismatch, then basically the neighbourhood will not establish.

Okay? So when you're designing, this is also an important thing we need to know, of course. Apart from the authentication, there are other parameters that should match. We'll see that in the troubleshooting section probably what the parameters are to match, but if there is an amismatch of areas on the opposite interfaces, then the routers will not form the neighbour relationship like here You can see this interface and this interface in area 0 and, of course, these two interfaces in the same area. And likewise, this interface is in area one like that, and again, you can see these two interfaces in area one here So the interfaces facing each other must be advertised or configured in the samearea. These are the five rules you need to keep in mind, so let me quickly summarise. You must have one area area zero compulsory. It means one area If you're using you can use any number 10.20.30 but when we use two areas, you have to use compulsory zero, so zero must be present and every nonbankbone area must connect to area zero, which means area one and area 51 cannot connect directly, they must connect viaone router which belongs to area zero. You must have one border router, compulsory. You can have more than one, but a minimum of one must be bought a router which is receiving the.

4. Multiple Areas – Configuration

Okay, so the next thing we'll try to see is the configuration example with the multiple areas. Like previously, we have seen the three out of three topology and we have configured all of them in one area, so I'll probably be using multiple areas here.

So before I go ahead, let me just quickly give an overview of the configuration. So we have already seen how to advertise these networks. If you remember, we have discussed the OSPF configurations. The first option is to go inside the OSPF and define the ProcessID. We have already discussed the concept of process ID, the basic configuration with singlearea added as your network. I want to order this subnet, so I'm going to specify ten to one, of course. I can be specific at 1055 if I'm giving the networkID, the wildcard mask, and the area and that interface, as per your design, goes in area zero, we need to advertise that particular interface in area zero. So when you're advertising this interface, let's say this 12 and 14 and as for my design, these interfaces fall into area 23, so when you're advertising those interfaces, we need to be specific and have to say that these two interfaces, of course here, these two interfaces in area four, you can see. So in terms of configuration, there is nothing different.

So The only thing left to do in the configuration is to define that specific interface, network, or IP address. IP means interface. Here in our case, it will interface false in which area, so you have to design accordingly. Like in general, as per the normal designing options, we do have a head office. Let's say I'm in Delhi and I'm connected to multiple branch offices in different cities. Let's say, let's say in hijabA. Let's say in Chennai. Let's say in Bangalore or Pune. different cities here and maybe you have multiple small branch offices, just like if you take any government office. Any government organization? They do have their own individual offices in small towns and villages, probably equipped with a small router, and you can group this as one area like that. So you can have multiple routers and they all connect to the main branch office in Hyderabad and they all connect here just like you can take an example of any passport offices or any different locations. You can take an example.

So when we design this general backbone network, we go with area zero, the backbone, including my head office along with all my branch offices. Because these branch offices will be connected to my backbone, the network backbone that is areazero, as well as my individual branch offices here, they will now act as myborders outers. So maybe they belong to area ten, let's say. So these all belong to area ten, including the interfaces facing the branch offices, so the interfaces facing the branch offices will be area ten, as will the interface facing my backbone. That is to my main head office, which may be in area zero, so it's all how you design, so you are not going to physically group the routers physically.

The way they are connected will be the same, but we are going to logically separate the different areas depending upon the network statements we use, like in this example. Like I showed you here in this example, The same thing here. You can see areazero connecting to non blackwood areas here. The configuration example is also the main part. The chain will be here based on which particular interface goes in which area, so the rest of the thing remains the same. The wildcard mark The processor remains the same and the interface goes into which area as per the design, like I said. This interface is facing towards the branchoffice, which will be in area 20. Let's say the interface which is facing towards the core will be in area zero, and this route is a borderrouter because it is going to multiple areas in general.

5. Multiple Areas - Configuration LAB

Okay, so the next thing we'll see is the configuration example here. Like previously, we have seen how to organize. So I'm going to use just three routers. Again, connecting 56 routers is really not possible in one lapse. So what I'm going to do is I'm going to just connect these three routers and I decided to go with my multiple areas. So when I do that, basically we have to make sure that we follow the rules. And in this design, I'm going to assign this interface to area zero because the rule says when you are going with two or more than two areas, we need to have one area zero. Of course, only one interface here, but production designs will not be the same as I said. And the next thing is I'm going to reassign this interface, this part in area ten. And the rule says that opposite interfaces should be in the same area, right? So these two are in the same area, area ten. And whenever I'm going with two or more than two areas, one should be zero, right? That's what we did.

And you must have one border auto. Now this is my border router area, the border router, which connects between area ten and area zero. And I'm also designing one more area here, this area, area 20. The same opposite interfaces should be in the same area. And area 20, when it is communicating with area ten, it should go via area zero, right? Of course, it is going via router two. Router two is a member of Area Zero, right? So you don't need to go wire zero, which means it's not like the middle of the area zero. It's not like that. It's going via any one router in area zero. So in our design, this router is part of area zero also, right? So it's a member of area zero. It's going to validate all the rules here. Or you can use this way, or you can just go and design this way. This is in area ten, this is in area twenty, and since this part is finished, I'll make it in area zero. So this also works. You can probably practise another scenario with this topology or with this design. So as long as it is going to make the rules, whatever the rules are, the Oscar is going to work.

Okay, so we need to make sure that whenever we have decided to go with multiple areas, it actually follows these rules here. So this is something that my topology. Let's go ahead and quickly configure this. And I do have this preconfigured topology here, and preconfigured means I do have the IP address preconfigured here. If you verify the ship's IP interface brief, you can see the interfaces are preconfigured and if you very refresh your IP protocols, I don't have any of the routing protocols configured here. So I decided to go with OSPF. So I'll start with the OSPF configuration. So I'll start with router one. So we'll say router OSP of one and I'm usingTen subnet, ten subnet which is on the van. Maybe you're using just one IP address. Then we can say ten to one. If you want to organise only one IP, we can use a wildcard mask area of ten. As per my design, ten networks are the interface and the 192 1681 network. Now, this network is probably in my LAN. So the submit mass is 24, which means the wild life mask will be 25 in area 10. So in my example, the router one, these two interfaces belong to area ten. So I have advertised them in the area Ten.

So let's go to router two. On the router two, if I go and check as per my router two configurations, I must configure this in ten because of opposite interfaces. Let's do that. So I'll say router OSP of one. When I say network ten network, you can identify that particular interface IP and then area area ten. So as for my example, it has to be area ten. So let's assume that we have configured the wrong area. Let's say instead of ten, if I say zero here. So I'll just try to show you that. So you should see some messages saying that the area mismatch happened. Of course, the neighbour ship will not form if you go and check Show IP with neighbors. Basically, the neighbourhood process is not going to start here. And let me see, the IP address of the interface is ten two. You can see on the router one it shows up like this is an area ten. This is zero. You can see it says invalid packet because there is a mismatch in area ID from the backbone area. I mean, you need to configure a virtual link if there is a mismatch. You can see these messages generally. So anyway, I just want to show you this. If you misconfigure the area, most likely you'll see those messages. Let me reconfigure this. I can reconfigure this to 10. So whenever you change, it will automatically change from zero to ten. That's what happens here. Initially, when I advertised, I advertised at zero. I've now changed them to ten. I can see the neighbour ship coming up immediately. Likewise, I'm going to advertise the other interface, the LANinterface, in which area Eleven dot network in area 20 and then my land interface 191 62 network. So we must have one area that is zero, right? At least one interface in area zero.

Now, we can assume that this interface goes and connects to my backbone router, maybe some other routers which belong to zero. So I don't have that design. But you can assume that in area zero. Done. So if you verify Show IPOs are ponderfaceinterfaces, or if you want to verify, we can use Showrun, probably Showrunning configurations. You can see this configuration. Of course, you can use the Show runpipe symbol and then section OSPO. So this works in the GNS or here. Basically, here, you can say begin option, begin with routers, and I think this option, begin with routers. It shows you the configurations with the router in your running configurations. So likewise, I'll quickly go to the router three and I'll also configure this on the router.

So let's confirm the interfaces are preconfigured with IPS and Router OSPF One. And I'm using the network 192-1683 dot network. As for my design, this three-dot network is in my area 20. And also the eleven dot interface, theinterface, which is in the van. So that is the only thing like this. So it is in area 20 R. So I should see the router two. And Router Three should also establish the neighbour relationship. If everything was perfect, Yes, you can see that. So verification is not different than what we did in our single area. So for the first verification, we always show the IP OSP of neighbors. We should see our neighbors' Radies and the States' full state timers and Active Interface IP. And if I say show IP route OSPF, the routing table will not change much. The same thing can be seen on the old routes. But one thing you'll see here, When I go to Router One, I confirm it with Router One. When I say, show IP route OSPO, Now here, you'll see the route as Oi.

So basically, there are two types of routes. "O" means the routes coming from the same area. And when you see OSPF interior routes, it means these routes are coming from a different area. So if you assume this is my topology, the Router One, Now, Router One belongs to Area Ten. And in this example, I'm receiving routes from two dot networks. The two-dot network now belongs to Area Zero, and these routes originate in different areas, correct? So these routes will be seen as OA OSPF interiors. Same thing. These three networks belong to Area 20, and these routes come from different areas because of area 20 routes. As a result, these routes will be regarded as OSP of Interair routes. So the cost values remain the same. The cost calculation is as we discussed in the OSP Metrics section. And similarly, if you go and check the routing table on the Router Two, on the Router Two, you will see these routes as zero because the Router Two belongs to Area Ten as well as Area Twenty as well as Area Zero.

So it's a bottle router. And the routes that are coming from 190 to 168 one subnet, are coming from Area Ten and are received by the router too. So it's area ten, area ten, right? So that's the reason you see O routes in a similar way. This three-dot network originated in Area 20 and was received there. So, which means the same area. So whenever routers are exchanging the routes between the same areas, then you'll see the output as O, and whenever you see them as OIA, it means you need to understand that those routes are coming from a different area and the rest of the options remain the same. And if you verify the database table of the routertwo, if I say Show iPOST database, we should see that the router two is a member of three different areas, so it is going to maintain the routes database or the links to the database of area zero. This is the area zero information, and it is from area ten as well, as it maintains information about area 20 also. But if you check on the router one, if I say Show iPOST database on the router one, now the router one belongs to area ten, so it's going to maintain the database of only the area ten advertisements. So it has LSS only coming from area ten and not from other areas. So the border router is responsible for receiving the illnesses from all the areas, and it's going to collect them all and maintain them in your links to the database. We can verify that by using the Ipospf database. Of course, if you want to check, the connectivity from end to end should work. But still, if you want to verify from the router one, I can go and pin to the router three gateway that should ping the router as well.

6. OSPFv2 - Interface Advertisements

Okay, the next thing we'll try to understand is the OSPF configuration with interface subcommentsor OSPF advertisements by using an alternate method. Like you remember in the previous concepts, we have seen the OSPF configuration. Generally, the Oscar configuration goes like this: when you want to advertise this ten network, we just define that particular network ID or the IP address if you want to order only that particular interface and then the wildcard mask, and then you have to tell that particular interface which area it comes in which area.So this is not the common way you say, because the first method initially we do this way, but the problem with this configuration method is that you need to really figure out, like in this topology, if I want to advise this particular interface in that particular area zero.

Then I need to determine what the interface IP address is, and then I need to determine whether you want to advise that particular complexsubnet or just an individual IP, because on LAN interfaces, we typically advertise only one IP address, which is sufficient to advertise the entire subnet. It's up to you. But again, when you talk about land interface, you need to reread the complete subnet again. So that means you need to figure out the network ID again if you want to ride as a complete subnet and then you need to calculate the wildcard mask if that is 27. Then you need to figure out the subnetmask and then calculate the wildcard mask. Define the wildcard mask and then the area number. As a result of the lengthy process, determining which interface is enabled with OSPO is difficult. Of course, we can use show Ipospo interface brief to figure out which interfaces we have enabled, but still while you are doing the advancements that will create a little bit of a complex way to do so, Cisco added another option with OSP of Vtwo OS per V Two, which means again for OSPO four IPversion four, we'll talk about OSPO V six.

When we say OSP of V three, that's what we call a task. So we have an alternate way to do this. So one option is you can still use this method. It's still valid, still correct. You can stilluse that method, but I can go with another way where I can simply use interface specific subcommandand I can just use that subcommand like IP, OSPO, something like that in the interfacemode instead of using the network commands. So we have an alternate way to advertise OSPF without using the network commands. So how to do that? We can directly enable the OSPF on the interface like in this example here. You can see let's say I want to organise this ten subnet like we are removing this ten network configuration, so I'm justreplacing this with the interface specific configuration like in this example. I'm changing this in configuration here. So if you just observe here, I'm going to zero by zero one and two, and these two interfaces should be in area zero. So we go to the interface and directly say that the interface should be advertised in area zero, so we don't need to bother about what the IP is because we are really not bothered about what is the subnetmask or what is the range of IP? This is not really required, but if you're using this method network statement, then you need to know the network ID. You need to know the wildcard mark and then define the area number, which is probably not required.

You just simply go and enable it under the interface and it will automatically advertise whatever the network is present with, of course whatever the network ID network present with. Whatever the mask you have configured on the interface will be advertised automatically to the remote servers. Likewise, here you can see these two interfaces. We have advertised them in area zero and then g zero byzero interface in area g zero by zero by zero. Again, we must examine g zero by zero. I think this interface has to be near23 as per this topology. Okay, so this is just an example. It is not exactly the same design topology here, but this is just an example to understand. So anyway, I'll be going into my specific topology. If you recall, in the previous lab I had already configured the OSPO, as I mentioned, show run, and you can go ahead and verify that the router one is preconfigured with this ten interface as well as this interface in both areas. So what I'll be doing is removing these configuration network statements and I'll be using the interface sub command.

So let's go and do this. I need to go to config mode and the first thing I need to do is remove this just to make sure that I already copied that and I'll also remove this. I'll copy this and then add no, or you can simply remove the entire OSP if you want. You can also go and say "all of us have one completely removed." Make sure that you enable the OSP process first, if required. You can also add an outrage if you want, even though routerid is optional as I said. If you don't add it, it will automatically take the highest type you have the loop back or the physical interfaces. We discussed this in the property sections, so we just need to enable the OS to process and then to advertise the network statements. We directly jump into the interface and then we say Ipospo, what is the process ID? I'm going to use one and that interface, which has an area of ten, and likewise, the other interface is zero by zero by zero.

So I also need to specify that interface in area ten. Now what I should see is I should see this interface and this interface forms enabled, of course, on the router too. I have configured the network statements. Here I have configured the interface subcommand, so it still works. So it's like alternate weight on this very fast show. The command supports OSPF interfaces, also known as OSPinterfaces. You can see these interfaces are advertised with OSP. Okay, so basically this interface sub command is an alternate way you advertise, and if you want to verify output why slightly, it will differ in the packet as well. It's not showing up, but when you work on GNS you can verify it. I'll show you these options here. When you say show IP protocols, if you areusing the network statements, then the output looks like this: the network ID, the wildcard mask in the area, and if you're using the sub interface interfacesub commands, then you see something like this. So all these interfaces are arodised in area zero and these are the interfaces which are under area zero.

So if you have more areas, it will also show you another area, and what are the interfaces under that? So, when you say show Ipospf interface, you will see these output differences. A specific interface This output is generally attached by an anetwork statement, which means this interface is advertised by the usingnetwork command. If you are using an interface subinterfacecommand or interface subcommand, then basically you see this option which is enabled means the OSPF is enabled on the interface and that is automatically advertising your interfaces in the OSPF domain. Okay? So either you advertise your OSPO, or this way is also correct by using a network statement. But if you're using these interface specific options, this will make your job much easier compared to figuring out the networkID and the wildcard master options. So, it will simply make your job much easier. So if you verify the route exchange, like if I go and check on the router too, you will see the same thing. You will see the routes one dot network is being advertised with.

Prepared by Top Experts, the top IT Trainers ensure that when it comes to your IT exam prep and you can count on ExamSnap Implementing and Operating Cisco Service Provider Network Core Technologies (SPCOR) certification video training course that goes in line with the corresponding Cisco 350-501 exam dumps, study guide, and practice test questions & answers.