00:00:00 - What better protocol to introduce first in this ICD 2 series
00:00:04 - than OSPF. And I say that only because OSPF is the
00:00:09 - most popular routing protocol in the world.
00:00:13 - OSPF is awesome in the sense that it is one of those protocols that
00:00:16 - just has so much complexity that as you move on into the CCNP
00:00:21 - track, dozens of videos will be dedicated just to this
00:00:25 - protocol and all of its concepts. But, it has been moved and introduced
00:00:30 - into the CCNA criteria because even in all its complexity
00:00:34 - it is the most popular routing protocol in the world. And by
00:00:38 - the time you're done here you will have a very good understanding
00:00:41 - of what it is and how to work with it. The CCNP will just build
00:00:44 - on that. It is kind of like you're.. You're walking up the mountain of
00:00:47 - OSPF and you'll get about halfway up in the CCNA level
00:00:51 - and CCNP just takes you to the peak of the mountain to where you
00:00:54 - know everything. So, in this video we're going to talk about,
00:00:58 - first and foremost, route summarization. The rest of OSPF
00:01:02 - won't make much sense without understanding route summarization.
00:01:06 - We'll then move into the OSPF terms and network design.
00:01:11 - Half of the battle in OSPF is understanding what these different
00:01:15 - terms mean, and why we design our network certain ways. We'll talk
00:01:19 - about that. Finally, we'll look at the OSPF 'hello packet,'
00:01:23 - which is the foundation piece of OSPF that allows routers to
00:01:27 - form neighbor relationships with each other and exchange routes.
00:01:31 - Many of the concepts that we talked about in OSPF will
00:01:34 - not make too much sense without understanding the idea of route
00:01:38 - summarization. Route summarization is all about making routing
00:01:43 - tables smaller. Here's the fact - the larger your routing table
00:01:47 - the more inefficient your router becomes. Your router is slower.
00:01:51 - And the reason why is because the router has more information
00:01:54 - to weed through for every single packet. It receives a packet and it's like ok, you're
00:01:57 - going to here, let me look at this routing table that's just massive
00:02:01 - now this, now this, it looks down through that whole routing table to
00:02:04 - try and find the best route. So what we can do to make our routers
00:02:08 - more efficient is to shrink the routing table. Summarization
00:02:12 - is how that's possible. Here is the idea. Let's say we have
00:02:15 - two routers in our organization. Well, let's say many routers
00:02:18 - but two in this picture. Router 1 over on the left and router
00:02:21 - 2 over here on the right. Now router 1 has a connection
00:02:24 - somehow to all these 192.168.x networks.
00:02:29 - We got 001020, all the way dah dah dah dah
00:02:32 - all the way down to 15.0. Now with every routing
00:02:37 - protocol router 1 is gonna send those routes over to router 2.
00:02:41 - That's what routing protocols do. So router 2 will now have
00:02:44 - all these routes in its routing table and it will say I can reach
00:02:47 - them, and that's how we know routing. That's how routing protocols
00:02:50 - educate each other. But, here's the problem with that. Router
00:02:55 - 2 now has 16 routes, 0 through 15, sitting in its
00:02:59 - routing table. That it.. Number 1 goes all the same direction
00:03:03 - to reach. There's only one way to get to router 1 and all of
00:03:07 - those separate routes, so, you know that's..that's the first..first issue.
00:03:11 - issue. The second issue is, if one of these networks go down,
00:03:15 - then that needs to be updated and replicated to router 2 and router 2 is
00:03:19 - gonna have to replicate that to other routers in the corporate
00:03:21 - network because they all have the same routes, and they all need to have
00:03:24 - a synchronized routing table. Now, honestly, when you will when
00:03:28 - we look at this picture does router 2 really need to know
00:03:33 - that the 1.0 network went down?
00:03:36 - Now, initially my thoughts go to well, yes it needs to, because
00:03:39 - maybe it has traffic that's going to that. Well, if router 1 knows that it's
00:03:43 - down router 2 will send a packet, the very first packet to router 1,
00:03:47 - and router 1 will then just reply and say ICNP unreachable.
00:03:51 - You can't get there from here. So, router 1 will take care
00:03:55 - of any packets that are going to that network, because it knows
00:03:58 - that it's down, so in the big picture, it really doesn't make
00:04:02 - too much sense for router 2 to know the specifics about those
00:04:06 - networks. So here's what we can do.
00:04:09 - Route summarization is the process of summing up all these routes
00:04:13 - into fewer advertisements. I'll give you what I will call
00:04:17 - cheap summarization. Here's an example. I can go on router 1, and I
00:04:22 - could say, I am going to advertise 192.168.0.0/16
00:04:33 - Meaning, I've pulled this subnet mask back to be a class B subnet mask.
00:04:37 - So, here's what in essence router 1 is saying, it's saying router 2
00:04:41 - I know about every single network that starts with
00:04:45 - 192.168. Router 2 puts that in it's routing tables and says,
00:04:49 - wow, that's a huge router over there. It's got all
00:04:52 - of my 192.168 networks. Now it only really has these
00:04:56 - 16 of them, but it's advertising the whole scope. Now when you
00:04:59 - do that, the routing protocol will automatically suppress all the more
00:05:04 - specific routes. So there's no need to go and send 192.168.0.0/24
00:05:08 - or 1.0 or 2.0 or 3
00:05:11 - because all of those are encompassed in 192.168.0.0/16
00:05:15 - It's the one big advertisement.
00:05:19 - Now that's.. that's like I said what I call
00:05:23 - cheap summarization, because
00:05:26 - it's not very efficient. Meaning, if I send that route advertisement,
00:05:32 - that means that this router, router 1, claims to have all 192.168 networks
00:05:37 - and I can no longer use 192.168
00:05:40 - networks anywhere else my network because router
00:05:43 - 1 has laid claim to them all. It..it says that I have them
00:05:46 - you cannot use them. So, it only has 16 networks but it
00:05:50 - claims to have them all, and we've now wasted a whole chunk of IP
00:05:53 - addresses that could be of use to us. So here's how you can do
00:05:57 - router summarization more efficiently. We need to get back to working
00:06:01 - in binary, and working in our subnetting mindset. Here's the idea.
00:06:05 - I've got all of these works. They all start with 192.
00:06:08 - They all start with 168. Now, we see the difference
00:06:13 - in the third octet. Let's break that into binary. 192.168.0
00:06:16 - is one, two, five, six, seven eight zeros,
00:06:20 - then all zeros over here, right? 192.168.1
00:06:23 - is 00000001 dot, and all zeros over here.
00:06:28 - Two, I'll just kinda, parenthesis here, is all zeros 1 zero. Three,
00:06:34 - is all zeros 1-1. Four is all zeros 1-0-0. Now we're all
00:06:40 - thinking in binary. And the, and the same trend goes down you know
00:06:43 - da..da..da..da..da. Let me get to, ah, I'm running out of space here.
00:06:46 - So, ah, let me, let me go to we'll say 13. Thirteen in binary
00:06:51 - is 128-64-32-16-8-
00:06:56 - 4-2-1, that's thirteen. Fourteen is 1110; Fifteen 1111.
00:07:03 - So, looking at all this, I know it's a little
00:07:07 - difficult to see because my binary's not perfect. I should actually,
00:07:10 - you know what, I should have typed these out. Hold on...
00:07:16 - Shazam. Take a look at that. Through the magic of video, I have already typed
00:07:21 - all of these up into binary. Now, let me go back to what I was talking about.
00:07:25 - We've got 192.168, and then here's our binary bits
00:07:28 - broken up, and I put the decimal number next to it. You see
00:07:31 - 0,1,2,3,4,5, and then down here I've got
00:07:36 - 13, 14, 15.
00:07:39 - Alright, the concept of route summarization is to take the bits that
00:07:44 - you have found that are similar, that are the same between
00:07:48 - all of these routes and group them together. So we look
00:07:53 - at our..our routes right here. We've got 192. That would
00:07:57 - be in binary, 8 bits that are all the same, right? 168,
00:08:00 - eight bits that are all the same, because every single one of
00:08:04 - these networks start with 192, and every single one starts
00:08:06 - with 168. Now we come to the third octet, and we look
00:08:10 - and we see, ok looks like, all those are the same, all these
00:08:15 - are the same, all these you can see my little lines going down; let me,
00:08:18 - let me, actually draw a perfect line right here, right...kachunk...kachunk.
00:08:24 - There. You see my magic dividing line? That is where the bits
00:08:30 - start to go different. So, what I could say is these have
00:08:35 - 8, 16, 17, 18, 19, 20. So, four extra bits here. Twenty
00:08:41 - bits that are all the same. So the perfect summary route that
00:08:45 - router 1 can advertise to router 2, is 192.168.
00:08:50 - .0,.0, slash 20.
00:08:55 - Notice I started with the very first
00:08:59 - network in my list. 192.168.0.0 so that that
00:09:04 - is going to be what I start with and 20. Now if you're thinking
00:09:07 - about this in terms of subnetting, if I had a slash 20, that
00:09:12 - would be 255.255.240.0,
00:09:16 - if I were to break this into binary my subnet mask would be
00:09:20 - 11110000. My increment, thinking
00:09:24 - back to subnetting here, would be 16.
00:09:27 - So, if I were to take 192.168.0.0 and I were reverse engineering this
00:09:31 - if you will to find out what my network ranges
00:09:34 - are, the first one would be 192.168.0.0. Second one
00:09:37 - would be 16.0, 32.0, and so on. So
00:09:42 - a /20 represents 0.0 through
00:09:46 - 192.168.15.255.
00:09:52 - Perfectly encompassing all of these routes that I have behind router 1.
00:09:55 - So router 2 now has the perfect route on its routing table.
00:09:59 - All of these can be suppressed. Now before I expand more on...on
00:10:03 - the specifics of the summarization, and kind of
00:10:06 - growing it a little bit;
00:10:07 - I want to mention what that does. Number one, it accomplishes
00:10:10 - our goal - larger routing tables equal slower routers, router 2
00:10:14 - now has a smaller routing table. Second, is it suppresses
00:10:18 - updates. If one of these networks go down, as I was mentioning
00:10:22 - before, router 1 no longer sends an update to router 2, and router
00:10:26 - 2, flooding the rest of the corporate network with that update,
00:10:28 - because router 2 doesn't care. It doesn't know 192.168.1.0.
00:10:33 - It just knows the big summary, so there's
00:10:36 - no purpose in sending the notification.1 is down,
00:10:39 - because router 2 doesn't really even know about
00:10:43 - the.1 network. It's all hidden. So we accomplish our two major objectives
00:10:48 - by putting that perfect summary route in there. Now, this
00:10:54 - summarization example
00:10:56 - is perfect. Meaning in the real world, I mean it would be awesome if
00:11:01 - router 1 had those networks behind it, but unfortunately, router 1
00:11:05 - or our organization, just went through a growth spurt, and they
00:11:08 - just added 192.168.16.0
00:11:11 - behind router 1. D'oh, totally goofs things up. Because 16
00:11:16 - in binary, if I were to put it in here, would be
00:11:20 - 00010000. Totally goofs up my summarization route.
00:11:26 - Now, some of you might be thinking, well, we can fix that. Right?
00:11:29 - We can, we can take that line of yours right there... Let's see if this
00:11:33 - works. Grab my line, and we could move it over. Look at that.
00:11:39 - The power of animation. We can move that line over and
00:11:43 - now you've got, that would be the first three bits..chum chum... are the same.
00:11:48 - So my new subnet mask would be
00:11:53 - 192.168.0.0/19 because we've moved our line back a
00:11:58 - bit to catch our extra 16 network that we added down
00:12:02 - here. That would be a solution and you're absolutely right
00:12:06 - in thinking that way that router 2 would now still only have
00:12:10 - one route on its table. But when you move that line back you
00:12:14 - didn't just catch the 16 route,
00:12:17 - you caught 17 (0001). You caught 18 (10010).
00:12:22 - You caught 19 (10011). Because
00:12:26 - all of those have the same three first digits, or three
00:12:30 - binary bits and the last five are different. So you've encompassed
00:12:35 - a lot more networks than just sixteen by doing that. Now that
00:12:38 - may be okay, but that's also I mean this this by the way
00:12:41 - will go down, dot..dot...dot, all the way to 32 because
00:12:45 - by moving our line back one, our new increment has become 32.
00:12:48 - So, we've encompassed networks all the way up to, I guess you could
00:12:52 - say, 31.255 would be the last network
00:12:56 - encompassed in this. 32 would start the next range.
00:13:00 - Now that's a lot of networks to do. So let me show you what
00:13:02 - most people will do in the real world. If you have growth, router 1
00:13:07 - will say, ok, I've got 16. I'm going to keep advertising
00:13:11 - that summary route this /20... sorry it's getting a little scribbly here...
00:13:14 - this /20 will be advertised to router 2, so it encompasses
00:13:19 - the first 15, and then I will advertise
00:13:22 - 192.168.16.0/24,
00:13:28 - as the separate route. So, in that logic, it's a lot better to
00:13:32 - have two routes in router 2's routing table than it would
00:13:35 - be have seventeen, 0 through through 16, so we're still
00:13:38 - accomplishing running efficiency, but we're not grouping a bunch
00:13:42 - of networks that router 1 does not have. So as router 1
00:13:46 - keeps growing, you know later on they add.17 and
00:13:49 - .18, and.19, those routes will be advertised individually
00:13:54 - until the organization reaches the point that they say,
00:13:57 - okay, we've got enough networks behind router 1 now. We can
00:14:02 - safely move that binary line back in our summarization to a
00:14:06 - /19, and encompass 0 through 31 networks,
00:14:11 - in that one summary route. So that is the idea of summarization and
00:14:15 - that key concept is what lights the way in OSPF, and why OSPF
00:14:21 - is such a powerful routing protocol.
00:14:24 - Now that you have the concept of summarization under your belt,
00:14:27 - we can move into the ideas and terminology behind OSPF.
00:14:33 - Half the battle in learning this protocol and this is a complex
00:14:36 - protocol, is understanding the terminology and the whats and whys
00:14:40 - that are used in OSPF, and the most foundation of all the terms
00:14:44 - is the concept of area. An area in OSPF is a group of
00:14:51 - routers that all have the same routing information. See here's the idea.
00:14:55 - When you have a network that continues to grow bigger
00:14:58 - and bigger and bigger, the routing tables on all the routers
00:15:01 - begin to grow bigger and bigger and bigger as well. So what
00:15:04 - we can do is split our network into groups of routers, like I can have
00:15:08 - you many routers within an area here.
00:15:12 - And all of those routers would have the exact same routing
00:15:16 - information. Here's a good analogy to describe it. In the trunk
00:15:20 - of my car I have a Rand McNally's road map of Arizona.
00:15:27 - Now also on the wall in my office right here, I have a world
00:15:32 - map. It's...it's actually an area code tracker that shows me area codes
00:15:36 - all around the world. And it's...it's this big world map that I can
00:15:39 - just look at and see area codes. Now if I were trying to figure
00:15:43 - out how to get to
00:15:46 - the mall, which is about 15 miles from my house. Would it
00:15:50 - be easier to use the Rand McNally road map in my trunk or
00:15:54 - would it be easier to use the world map sitting right here on
00:15:57 - the wall?
00:15:58 - Well, it would be easier to use the one in my trunk because it's focused
00:16:02 - on specific areas of Arizona. And instead of looking at the
00:16:06 - world map and going, man, all these roads are so small. I can't...I can't see
00:16:10 - you know, could I use the world map? Yeah, maybe, you know if
00:16:13 - I can look close enough, and they actually put the road small enough
00:16:15 - on there. I might be able to do it but it would be a lot harder because
00:16:18 - there's so much more information I have to weed through to get there.
00:16:21 - But the one in my trunk is just more focused.
00:16:25 - That's the idea of areas. Once your organization grows too big
00:16:30 - all of your routers will have to process all that information.
00:16:33 - And every packet that they get, it's like they're looking at a world
00:16:36 - map of your organization. And it's going to slow them down. So, by breaking
00:16:39 - it into areas, you could say, okay well area 0 represents
00:16:43 - we'll say Arizona. Area 2 represents Florida. Area 1 over here
00:16:48 - represents California and the United States. And we have these
00:16:51 - specific areas that group together similar routers.
00:16:55 - Now, just to give you a random guideline. CISCO recommends that
00:16:59 - an area never be more than 50 routers.
00:17:03 - So, as your network grows, you can begin dividing into areas.
00:17:07 - Now, that is...that is a guideline. That is not a hard and fast
00:17:10 - rule. So now that we see what areas are, let's talk about the
00:17:15 - routers that make-em up. Inside of the areas, and I'm gonna stray
00:17:19 - straight from talking specifically about the backbone and types
00:17:22 - of areas and stuff like. I'm gonna first talk about the routers.
00:17:26 - Inside of an area, you'll have internal routers. And these are routers
00:17:30 - that belong to an area. That internal router connects to area
00:17:33 - 0 and knows nothing but area 0. In area 2, I have
00:17:37 - another router. That router is an internal router. In area 2, and
00:17:41 - it's...it knows nothing but area 2. These routers that sit
00:17:45 - between areas are known as ABRs. Area Border Routers.
00:17:52 - Now these are usually the beefier routers in your network. The ones
00:17:55 - that have a little more processing power a little more memory
00:17:57 - than the rest, because these routers have the road maps
00:18:02 - for two or more areas in the routing table, so they have to be
00:18:05 - able to process and look through, it's almost like two page
00:18:08 - of pages of the map, rather than one page. The big point about
00:18:13 - an ABR that
00:18:15 - you'll want to know, is that an ABR is the one that is able
00:18:20 - to summarize. Summarization. Now you know why I covered that
00:18:27 - key concept on on, a page ago because if you didn't know
00:18:30 - what summarization was all about, the whole design of OSPF wouldn't
00:18:33 - make any sense. And when you're designing these areas, it has to be
00:18:38 - a hierarchical design. And what that means is that you group similar
00:18:43 - subnets in similar areas. So, for example, in area 1 you
00:18:47 - know I've got my 50 routers, and maybe over here I do
00:18:50 - 172.16.1,.2,.3, all of these
00:18:53 - different subnets, all /24s we'll say for ease.
00:18:57 - And I've got all these different subnets. And the ABR
00:19:00 - as it advertises area 1 to area 2, can sum that up and say, oh
00:19:05 - area 1 is all about 172.16.0.0/16.
00:19:09 - Yes I'm using some cheap summarization,
00:19:13 - just for easing here, some easy stuff. But at the same time,
00:19:16 - you get the concept. The area 0 backbone routers, they don't
00:19:21 - need to know anything more about area 1 then that one route.
00:19:25 - So 50 plus routes summed up in one. Same thing with area
00:19:28 - 2. Maybe this is 172.17.1,.2,.3, and so on...
00:19:33 - And we sum that up as it comes in the backbone. And the backbone
00:19:36 - has a hierarchical network of its own. If you don't design your
00:19:40 - network right with OSPF, it will tear you apart, because the ABRs
00:19:45 - will not be able to summarize, and there's no point in
00:19:48 - dividing into multiple areas. And let me emphasize that. The
00:19:51 - whole reason that we even use multiple areas is to summarize.
00:19:57 - If you don't summarize when you break into multiple areas,
00:20:00 - you're defeating the whole purpose and you're just causing more
00:20:02 - processor cycles on all of the routers. So, when you're setting
00:20:06 - up an OSPF network, you have to be very careful where to place things.
00:20:10 - If I were to go to area 1, and just say, ah, ah, let's you know let's 192.168.1 over here.
00:20:14 - Ah, let's throw that.10 network...
00:20:17 - over there. You're...you're shooting yourself in the foot, because you
00:20:20 - can't summarize that in a hierarchical format. That's known
00:20:24 - as an IP address hierarchy. So, let me hit the specifics that I have
00:20:29 - here. All areas, the rule of OSPF, all areas must connect
00:20:33 - to area 0. That's what's so special about area 0. It is considered
00:20:38 - the backbone of your network and all other areas must connect
00:20:42 - over here as your network grows larger and larger, things must
00:20:46 - tie into that area. All routers within an area have the same
00:20:50 - topology table. And if you want to emphasize that, that means
00:20:54 - road map. All of the routers in area 0 know everything
00:20:59 - about area 0. Even the routes that they're not using. The backup
00:21:02 - routes, you know, and they know everything. And that's fantastic
00:21:05 - because if a route goes down in that area, the routers, wham
00:21:09 - converge in a snap. They're able to find those backup routes
00:21:12 - that they have in their road map. They just pull the road map back out and
00:21:15 - regenerate the routing table. Now let me emphasize the
00:21:18 - difference here. All routers in the same area
00:21:22 - have the same topology table. Or they all have the same roadmap.
00:21:25 - But, every router within the area will have a different
00:21:32 - routing table.
00:21:34 - Hmm, let's talk about that. We've got, we'll say, this...this router
00:21:38 - up here. This area 0 is represented as Arizona. Maybe this router
00:21:42 - is what connects to Phoenix. This router over here connects to
00:21:45 - Tempe, that's another city here. And this router over here connects to Tucson.
00:21:51 - So we've got, you know, these...these three different routers. Now, in
00:21:54 - Tempe it'll have the road map of the entire area 0, the entire
00:21:58 - backbone. Tempe knows how to get to Tucson. It knows that it
00:22:02 - has a backup route to go to Phoenix to go to Tucson, but it can
00:22:05 - also go to whatever the city is. Scottsdale we'll say. And get to Tucson. It's got
00:22:09 - all this information so if its primary route through Scottsdale
00:22:13 - fails, it just looks back at the road map and says oh, I've got another
00:22:16 - route through Phoenix. That's meaning the same topology table. All
00:22:20 - routers in...in area 0 will have a different routing table
00:22:23 - because they all start from different points.
00:22:26 - So, Tempe will say, my best route to get to Tucson might be
00:22:31 - through the Scottsdale router. Whereas Phoenix will say well I've got a
00:22:34 - direct connection to Tucson down here. I'm going to use that as my best
00:22:37 - route to get to Tucson. So even though they have the same road
00:22:40 - map, they all generate different routing tables. It's just like,
00:22:44 - I mean think about it logically. If you have the same road map
00:22:46 - in your trunk that somebody else has and they live some completely
00:22:50 - different place than you, their best routes around the state
00:22:53 - are going to be different because they're starting point is
00:22:55 - different. They...they are in a different place in the
00:22:58 - state, so they will generate a different routing table.
00:23:02 - The goal of OSPF is to localize updates within an area.
00:23:06 - Whenever something happens in area 0 everybody knows about it.
00:23:09 - But,
00:23:11 - area 1 should not. Because we should be summarizing to area 1.
00:23:16 - And area 2 should not. So only things that happen in area 0
00:23:20 - will stay in area 0. It's like our Las Vegas mantra.
00:23:23 - What happens in Las Vegas, stays in Las Vegas. Same thing here.
00:23:26 - What happens in an area stays in an area. Finally, and I talked about this;
00:23:31 - requires a hierarchical design. You must design your network right.
00:23:36 - Oh, there is one more thing. Let me erase all this chicken scratch. The last
00:23:40 - term I want to throw out at you
00:23:42 - is this one over here tucked in the corner. The
00:23:45 - Autonomous System Boundary Router or ASBR. The Autonomous System Boundary
00:23:50 - Router is the router in OSPF that connects to networks outside
00:23:55 - of its own. This is not another area. This is a completely different
00:23:59 - network. So it could be a network that is running Rip over here.
00:24:03 - It could be the internet and I would say that's the most common
00:24:06 - network that the ASBR connects to. Others...there's many different
00:24:09 - things that...that this could be. But the ABR and the ASBR are
00:24:14 - the only two routers in OSPF that can do summarization.
00:24:18 - Between areas and between completely different routing systems.
00:24:23 - The last OSPF concept I want to talk about is how OSPF
00:24:27 - forms neighbors. Unlike the Rip protocol, OSPF will form
00:24:32 - direct neighbor relationships with the routers it wants to
00:24:35 - speak with. Rip just walks up to the ethernet line and says, hello
00:24:39 - everybody, sends out a broadcast to everybody. These are the routes
00:24:42 - I know about. It doesn't actually form neighbor relationships, it's just
00:24:45 - the other routers happen to hear it saying hello everybody,
00:24:48 - and adds those routes to its routing table. They don't know about
00:24:51 - each other directly. So, in OSPF, routers come up to each
00:24:56 - other and say, hello router, hello, and they start exchanging
00:25:00 - routes between each other and then they maintain that...that
00:25:02 - neighbor relationship using something known as the 'hello protocol.'
00:25:07 - It's not just what I'm saying. That's the technical name of the protocol. Now
00:25:10 - hello messages are sent, when you configure OSPF on whatever
00:25:14 - ...whatever interfaces you designate. So, if I say head out serial
00:25:18 - zero zero then it will start saying hello and trying to form
00:25:21 - neighbors on that interface. If it does, these neighbors will
00:25:25 - meet and they will exchange routes and now we have a synchronized
00:25:28 - routing table. In OSPF, these hello messages are sent
00:25:32 - once every ten seconds on broadcast or point to point networks
00:25:36 - And once every thirty seconds on non-broadcast multi-access
00:25:41 - networks. Those are things like frame relay which we'll talk
00:25:43 - about later. The idea is that the more often you send hello
00:25:48 - messages, the sooner you will know if a neighbor is down, because
00:25:52 - they'll stop responding to your hellos, and the faster you
00:25:54 - can change over to a backup route.
00:25:57 - Now, a lot of people in the OSPF world will tune this
00:26:01 - hello timer down, to a second or maybe two seconds. So you're just
00:26:05 - sitting there going, hello hello hello, making sure that they're
00:26:09 - online because you want to be able to detect that failure extremely
00:26:12 - fast. Now, when you and I say hello we think of a greeting.
00:26:18 - Like, hello how are you doing, or...or something to that effect. But
00:26:21 - when OSPF sends a hello, I want you to think about it like
00:26:24 - an envelope with hello written on it. And when that hello message comes
00:26:29 - across, the router opens that hello envelope and sees all of these specifics
00:26:34 - inside of it. It will see things like the router ID, which
00:26:38 - is the name of the OSPF router over here. It says hello, my name
00:26:42 - is, and the router IDs is an IP address. It might say 184.108.40.206
00:26:44 - and the router will say, well hello, I'm 220.127.116.11
00:26:48 - thats...thats the router ID. In that hello envelope
00:26:51 - will be that hello and dead timers, meaning how often they're
00:26:54 - saying hello, and it's kinda rude, but that's alright. It's how...how soon
00:26:58 - until they believe you are going to be dead. You know how many
00:27:01 - hellos they can miss before they say that person must be down.
00:27:05 - They will advertise their subnet mask in that hello packet.
00:27:08 - They will advertise what area they are in in that hello packet.
00:27:11 - Now you notice some of these
00:27:15 - messages or I should say pieces of that hello envelope have
00:27:20 - little stars by them.
00:27:22 - The stars, I'll put a little key, star equals must match.
00:27:31 - Right? They have to match in the hello packets between the neighbors,
00:27:36 - or else these guys will not end up forming neighbor relationships.
00:27:41 - I think about it this way. My old roommate that that I used
00:27:44 - to live with. Ah, actually met his wife online. He...he had one
00:27:50 - of those
00:27:51 - dating, um, dating sites. I don't know what they're called,
00:27:55 - where you...you match up. Right? And when you go to this dating site...
00:27:59 - I actually was working with him because I'm the computer
00:28:02 - guy of the house. And I was showing him how to get on there, and you
00:28:05 - know, log in and all that. And, ah, on this dating site you have
00:28:08 - criteria. And what you do in your criteria, you say,
00:28:11 - okay, this must match. For instance, you know this...this...this person
00:28:16 - or this mate that I'm looking for must have blue eyes.
00:28:20 - They...they must have an adventurous spirit. Meaning they like
00:28:24 - to do adventurous things. They must have, you know, and you list
00:28:27 - your must-haves. And then you list your, well you know, it would be nice if
00:28:31 - they had, um,
00:28:34 - pink painted toenails. I don't know. I'm just throwing stuff out there.
00:28:37 - You know. But you know if those don't match it's okay. In the
00:28:39 - same way, our routers are online daters. They are sending
00:28:43 - their little hello messages with each other and inside
00:28:46 - of there are criteria that must match. For instance, if this
00:28:50 - router 2 over on the righthand side says hello once a second and
00:28:54 - the one over on the left says hello once every ten seconds then
00:28:58 - it's gonna say, I'm...I'm sorry. We don't match. We're not compatible
00:29:03 - with each other. And at that point they will choose not to form
00:29:07 - a neighbor relationship. Now, a lot of these things that you
00:29:10 - see in this hello packet we haven't talked about yet. And we will talk
00:29:13 - about but the ones with stars, if they're not matching, the
00:29:16 - neighbor relationship will not work. And that's the number
00:29:19 - one trouble shooting criteria we have with OSPF, is
00:29:23 - making sure that all these things match. Otherwise routes won't
00:29:27 - be exchanged.
00:29:29 - That should be enough of the OSPF concepts to get us going.
00:29:33 - As we continue this in the next video we'll be able to see
00:29:37 - how these concepts apply in our configuration. But before we
00:29:41 - do let's wrap things up here. We first off looked optimization
00:29:45 - at its best, or the best way to optimize a router is through
00:29:48 - route summarization. We then moved into the OSPF terms and
00:29:52 - network design. And to hit the high ones, area is the most critical. Area
00:29:57 - defines routers that have the same topology database or road
00:30:01 - maps. ABRs, Area Border Routers, is what moves you between areas. And
00:30:06 - ASBRs, Autonomous System Boundary Routers, are the routers
00:30:10 - that move you outside of your own OSPF network, maybe to access
00:30:14 - the internet. We've then finally analyzed the OSPF hello packet.
00:30:20 - The hello packet is the foundation language that these
00:30:23 - these routers will use to communicate with each other
00:30:26 - and form neighbor relationships. If the relationships don't form
00:30:30 - routes won't be exchanged. I hope this has been informative for you and
00:30:34 - I'd like to thank you for viewing.