00:00:00 - Now we'll begin our next major section of technology, which is routing
00:00:05 - protocols at the ICND2 level. Back in ICND1,
00:00:08 - all we really discussed was RIP because that's the basic
00:00:13 - routing protocol you can use to get a small network running.
00:00:16 - As we move into ICND2, we'll move in to some of the more
00:00:20 - advanced protocols such as OSPF and EIGRP. But
00:00:24 - before we begin, we need to understand some foundation concepts
00:00:28 - and hit a little review from ICND1 as we discuss distance
00:00:32 - vector versus link state routing protocols. This is a big
00:00:37 - class difference of routing protocols. Distance vector was what
00:00:40 - we saw on the ICND1, and now we're going to move into link
00:00:43 - state in ICND2.
00:00:45 - We'll first look at what distance vector routing protocols are
00:00:48 - all about and talk about some of their drawbacks.
00:00:54 - When we were in ICND1, we talked about RIP, and we
00:00:58 - said this is what it does, but we really didn't see the dark
00:01:02 - side of the RIP protocol. That's what we'll look at as we see the loop
00:01:05 - prevention mechanisms. We'll then move into the link state routing
00:01:09 - protocols and what they're all about.
00:01:12 - We'll first start off with a little bit of review from the ICDN 1
00:01:15 - world, as we talk about the two umbrellas of routing protocols
00:01:20 - that exist: distance vector and link state. These two umbrellas
00:01:25 - really mark a big divide between the styles of routing protocols,
00:01:29 - and the best way I can compare them is to talk about my first
00:01:32 - car. My first car was a 1982 Volkswagen Rabbit diesel model.
00:01:40 - Now, nowadays, I look back and I think, man, I wish I had that car because
00:01:44 - gas prices are so high, and I think I got around 50 miles
00:01:46 - to the gallon. It was insane. But back when I was 16 years
00:01:51 - old and in high school, I didn't appreciate the car. I thought,
00:01:55 - this car is not a babe magnet, because that's, that's
00:02:00 - you know, admittedly, that's all you're really looking for at the high
00:02:03 - school-age, is just the cool car. And that car lasted me for years.
00:02:09 - As a matter of fact, I sold it when I had 200 and
00:02:11 - 200,000 miles on it, and it was still running. I never
00:02:15 - changed the oil; I just put gas and I never did any maintenance
00:02:19 - at all. It just kind of went. The drawback to the Volkswagen Rabbit,
00:02:24 - besides the babe magnet factor, was the speed. I timed
00:02:30 - it once. You know to a high school male
00:02:35 - person, as myself, the 0 to 60 time of your car is very
00:02:40 - important - to see how fast you can move. And I timed it, and that
00:02:44 - car got 0 to 60 in 42 seconds. I still remember
00:02:49 - it; to this day I thought, how horrible! But that's what distance vector
00:02:54 - routing protocols are all about. They're extremely easy to configure.
00:02:58 - Meaning, just like my car, it didn't really take much to get
00:03:01 - in there. Although, I had to learn to drive stick shift and what
00:03:03 - glow plugs were all about. You just kind of turn them on, and they work
00:03:07 - and you don't need to maintain them - they just keep running. The
00:03:10 - problem with them is they're not very fast, and they don't
00:03:14 - have many features. The two examples of distance vector routing
00:03:18 - protocols are RIP and IGRP. The RIP protocol we set
00:03:22 - up in ICND1. IGRP is gone. CISCO has officially
00:03:29 - discontinued support for IGRP, it's not even in the newer IOS
00:03:33 - versions. That's because nobody really used it.
00:03:37 - They moved on to link state. Now, that is the Dodge Viper,
00:03:41 - the screaming fast car that takes just a ton of cost
00:03:47 - to get running, meaning a lot more knowledge to run that car than just
00:03:51 - a Volkswagen Rabbit, and there's a lot of maintenance and tuning and tweaking
00:03:54 - that has to go into it. And in the same sense, with link state routing protocols
00:03:59 - they're very difficult to configure. As a matter of fact, we're
00:04:02 - going to get into the OSPF protocol and talk about configuring
00:04:06 - and we'll even set it up, but to really fully understand OSPF
00:04:12 - that's what the CCNP courses are all about. The BSCI
00:04:15 - focuses tons of information on OSPF that expands on what
00:04:21 - we're going to lay as the foundation here. So they are feature-riffic;
00:04:24 - they have a lot of stuff that they're able to do and a lot
00:04:27 - of speed criteria that they can handle. And the two examples
00:04:31 - are OSPF and IS-IS. IS-IS was the routing
00:04:36 - protocol for the OSI protocol. A lot of people just think OSI
00:04:41 - was a model, but there is an OSI protocol that is out there
00:04:44 - and it's even better than TCP/IP, but not many people
00:04:47 - use it.
00:04:49 - Now down at the bottom, you see the HYBRID and this is the Dodge
00:04:52 - Viper Rabbit if you will. It's the best of both worlds, really
00:04:56 - easy to configure but gives you all the features of a link state
00:05:00 - protocol. The Achilles' heel of this hybrid routing protocol
00:05:05 - is that it is proprietary, and the one example of it is EIGRP.
00:05:09 - CISCO made it and only CISCO runs it, so you have to have CISCO
00:05:14 - routers everywhere to support that protocol.
00:05:17 - Let's first focus on distance vector routing protocols in their
00:05:20 - simplicity and in such a way that we can see some of the weaknesses
00:05:25 - that were not exposed in ICND 1. Distance vector routing
00:05:30 - protocols by nature send their entire routing table at specific
00:05:34 - intervals. The one protocol that's left, which is RIP,
00:05:37 - sends its entire routing table once every 30 seconds. So
00:05:41 - you can think of it as - imagine, imagine yourself standing
00:05:45 - in front of a big room of people,
00:05:48 - and you step up to the microphone every 30 seconds and
00:05:51 - say, "Hello everybody in this room, I know about
00:05:56 - 192.168.1.0,2.0,3.0,4.0." And then you go and
00:06:00 - take a seat and wait 26 more seconds and
00:06:03 - walk back up to the microphone, (it's been that 30 second interval), and you
00:06:06 - say, "Hello everybody, I know about 192.168.1.0,2.0,3.0,4.0."
00:06:10 - And so on, and then you sit back down. That's
00:06:13 - exactly what RIP is doing. It's broadcasting or multicasting,
00:06:17 - depending on the version you're using, to the entire network once
00:06:21 - every 30 seconds to let them know what it knows about.
00:06:24 - So if we're running RIP down here, I have
00:06:28 - a interval once every 30 seconds, I'm saying, hello everyone
00:06:32 - out both interfaces, I know about 192.168.1.0,2.0,3.0 and 4.0.
00:06:36 - Now, this router over here gets it, and says, oh
00:06:41 - great, well I knew about two, but I didn't know about one, so I'll add that
00:06:44 - to my routing table. It's a little review in, in, from ICND 1, how routing
00:06:49 - tables are built. And so, before long all these routers know
00:06:52 - about their routing table. Now keep in mind, the reason that
00:06:56 - RIP sends updates every 30 seconds
00:07:00 - is not just because these routers might have missed something,
00:07:04 - it's because its update system is the only keepalive it has.
00:07:09 - Meaning, if this router stops hearing updates it's going to
00:07:12 - realize, oh well, this router over here, (we'll call them router C),
00:07:16 - is dead. So, I will see if I have backup routes to reach
00:07:20 - the 192.168.1.0 network that router C was
00:07:24 - being used to support. So, those updates are keepalives, and
00:07:28 - in their simplicity, RIP or distance vector protocols
00:07:33 - have looping issues. Here's what that means:
00:07:38 - Let's imagine that router C over here has, you know, sent out
00:07:44 - its update. It's doing the normal RIP thing, and, you know, over
00:07:48 - here is the building that it's supporting, we'll say it's in Arizona,
00:07:52 - and that's the network. And the IT admin in Arizona comes
00:07:57 - in. It's seven in the morning, you know, he just happened to beat rush-hour
00:08:00 - so he came in a little early. And that, that, the administrator is
00:08:03 - walking up into the IT room. Just got his cup of coffee, you know, (slurp), sipping the
00:08:08 - coffee and just you know, trying to wake up. Network admins
00:08:13 - are not morning people. They are creatures of the night because that's
00:08:16 - when most of your maintenance Windows are happening. And he
00:08:19 - walks up, (trot,trot,trot,trot,trot), sees this dangling cable,
00:08:24 - but sees it too late. Trips over this Ethernet cable connecting
00:08:28 - the router to the switch and the rest of the network over here.
00:08:32 - And you know, severs the cable, (snap!), and sparks are flying
00:08:35 - and fire extinguishers. It's not that bad. But, either way,
00:08:40 - router C loses its connection to the 192.168.1.0
00:08:44 - network. Now, unfortunately, router C just sent an
00:08:48 - update two seconds ago, so it's got 28 more seconds to go
00:08:53 - before it sends out its next update and lets everybody know
00:08:56 - that the network is down. Well, here's where the problem comes
00:08:59 - in. Router B just happened to send out its last update five seconds
00:09:04 - or sorry I should say, 25 seconds ago, so it's got five seconds
00:09:09 - left until its next update. Four..three..two..one.. "Hello everybody,
00:09:15 - this is router B now, I know about 192.168.4.0,3.0,2.0
00:09:19 - and 1.0, (chrk,chrk,chrk,chrk,chrk). That update hits router C. Router
00:09:25 - C gets it and goes, "That's, that's fantastic. Router B, you must
00:09:29 - have been reading my mind because I just lost my ow- my own
00:09:33 - connection to 192.168.1.0. It just went down.
00:09:37 - So, what I'll do is I'll point to you as the next hop address.
00:09:43 - We'll say router B is 192.168.2.2 over
00:09:46 - here. So router C will point to router B as the next hop
00:09:50 - address to now reach 192.168.1.0.
00:09:54 - Now, you and I are looking at this diagram going, why are you doing
00:09:59 - that router C, what are you thinking? Don't, no, don't do that.
00:10:02 - Router B was using you to get to the 192.168.1.0 network.
00:10:05 - Why, why would you try and use router B? But, unfortunately,
00:10:10 - distance vector routing protocols aren't that smart, so router
00:10:13 - C points to router B to reach the 192.168.1.0.
00:10:16 - network, and 28 seconds later sends
00:10:21 - out its own update saying, Hey, I now have a new path to get
00:10:24 - to 192.168.1.0. It is through router B.
00:10:29 - That's, that's the way that I go, and it's two hops away from
00:10:34 - me. Now, let's check this out. Before, router C was directly
00:10:39 - connected, so it was zero hops - that's RIP's metric, it's, it's hop count.
00:10:44 - So it was directly connected zero hops. So for router B, it was one hop, meaning
00:10:50 - router B would go- ta,ta,ta,ta,ta- hop, router C and I'm there. That, that
00:10:54 - was the process router B would use, but now router B sent
00:10:58 - an update to router C saying
00:11:01 - I have a link to 192.168.1.0, and it's one hop
00:11:05 - away from me. Router C got that update and says, well, if it's one
00:11:08 - away from you, and I go through you to get there, it must be
00:11:12 - two hops away from me.
00:11:14 - Does that make sense? Because router C thinks, I hop through router B
00:11:17 - and then go wherever router B goes and then I'm there. So router B
00:11:21 - gets the update, right? Says, Ok, now I am hearing about
00:11:25 - 192.168.1.0 from C. And C is
00:11:28 - now claiming that it's two hops away. Router B is scratching its
00:11:32 - head and goes, well, that's funny. It used to be zero hops
00:11:36 - away from router C, but now it's two, so, well, I guess if it's
00:11:40 - two hops away from router B, it must be three hops away from
00:11:43 - me, because I use router- oops, I meant router C- if it's two hops from router
00:11:48 - C, it must be three from me, because I use router C to get
00:11:51 - there. That's odd. And it passes that update to router A over here,
00:11:56 - and router A gets it and says, well, that's funny, it used to be two hops
00:12:00 - for me, but I guess now that it's three hops from you, it must be four hops from me. And
00:12:04 - this system goes round and round and round and round and round.
00:12:08 - And these hop counts keep going up, higher and higher
00:12:10 - and higher. This is known, its symptom is technically called
00:12:14 - a countdown to infinity. Now, you and I know you'll never reach
00:12:18 - infinity, and that's why we call this an official routing loop.
00:12:22 - This is something that distance vector routing protocols, such
00:12:25 - as RIP, can experience.
00:12:28 - Thankfully, built-in to CISCO routers are five loop prevention
00:12:33 - mechanisms to keep what I just showed you from happening. Now
00:12:38 - this will prevent every routing loop that you can have. So
00:12:41 - that leads to the question, or begs the question, why do we need to
00:12:45 - know about these? Well, the reason why is they do help stop routing
00:12:49 - loops, but they can also cause problems. And there may be times
00:12:53 - where you have to turn one or two of them off. So, the first
00:12:58 - loop prevention mechanism that we have is a maximum distance.
00:13:01 - With RIP, the maximum distance that's defined is 16,
00:13:05 - I'm gonna grab my pen here, 16 hops. So, once a RIP network is 16
00:13:12 - hops away, it is then considered dead. So, if we had a loop
00:13:17 - that did happen in router.. Let me put my letters back here C, B and A
00:13:20 - were passing that 192.168.1.0 network
00:13:24 - all around the network, that would keep going until they reach
00:13:27 - 16 hops, and then whichever router got it at the 16 hop level would
00:13:31 - say, oops, that route is now dead; it's too far away. That also tells
00:13:36 - you that RIP can only be used on small networks, because there
00:13:39 - are some networks that have links that are farther than 16
00:13:42 - hops from each other.
00:13:45 - The second one that you see is route poisoning. Route poisoning
00:13:50 - kind of integrates with the maximum distance. And what it does
00:13:54 - is advertise that the network is down immediately. So it seems
00:13:59 - kind of funny, but routing protocols built-in themselves do
00:14:03 - not have a way of saying, this route is down.
00:14:07 - That's not what they're designed to do. Routing protocols are
00:14:10 - designed to tell each other what routes are up. So route poisoning
00:14:14 - had to be specially engineered to say, Router C, as soon as
00:14:18 - this network goes down, poison it, kill it. And the way it kills
00:14:22 - it is by setting it to a maximum hop count. You know it kind of
00:14:26 - triggers another rule. These, these rules integrate tightly
00:14:29 - together. So as soon as that network admin trips over the wire,
00:14:32 - over here in Arizona, it
00:14:35 - poisons the route and says, well, this is now 16 hops away, and the
00:14:39 - next update that goes out will advertise that. So router B
00:14:42 - gets it and says, oh, well that's the maximum, so that route must
00:14:45 - be dead. Now, the third one is just logical. One of the reasons
00:14:51 - this whole system happened, was router C sent an update two
00:14:55 - seconds ago, right? And so when this network went down it had
00:14:58 - to wait 28 more seconds before the next update went
00:15:02 - out. And with that delay, router B was able to send an update
00:15:05 - and say, hey, router C, I've got a route to 1.0 and that caused
00:15:09 - the whole problem. So triggered updates integrates with route poisoning, which integrates
00:15:14 - with max distances, which says when that network admin trips
00:15:17 - over the wire, immediately, immediately negate all timers.
00:15:22 - All bets are off at that point. You've had a major network change.
00:15:26 - So at this point, we will trigger an update; overrule that 28
00:15:30 - seconds you were going to wait, router C, and as soon as you
00:15:33 - see the network go down, send a poisoned route, telling router
00:15:36 - B the network is down.
00:15:40 - Split horizon, split horizon is one of the most difficult to
00:15:45 - remember, but one of the most critical ones, and one of the ones
00:15:49 - that causes the most problems when, when you are using these
00:15:52 - loop prevention mechanisms. If there's one that you're going to disable
00:15:55 - it's probably split horizon. And when we get to some of the advanced
00:15:59 - WAN networks I'll bring this back up. What split horizon does
00:16:03 - is tell routers: Do not send updates back in the same direction
00:16:08 - you receive them on networks that are being advertised. Let
00:16:13 - me explain that in plain English.
00:16:17 - Let's say that you and I met on the street. And you came up
00:16:22 - to me and said, "Hi Jeremy, my name is Mike," and I said, "Hi Mike." And,
00:16:26 - and you said, "Jeremy, I have brown eyes."
00:16:32 - Now, would I ever tell Mike that he has brown eyes? No, because that
00:16:37 - would create a people loop. Meaning, if, if Mike and I were to
00:16:41 - meet on the street, and Mike said, "I have brown eyes," and told
00:16:45 - me about that, I would never tell Mike he has brown eyes, for many
00:16:48 - reasons, but primarily because I don't want to start a people
00:16:52 - loop because I would say, "Well, yeah, you have brown eyes," and Mike would say,
00:16:55 - "Yeah, I do," and I would say, "No, you have brown eyes." And we'd just keep going around
00:17:00 - and around until both of us, Mike and I, died. So split horizon
00:17:04 - is a rule that says, do not tell routers about routes they told you
00:17:09 - about. So when router C says, Hey router B, I know about 192.168.1.0/24.
00:17:13 - Router B is now banned
00:17:18 - from telling router C about that route. Brilliant! Because
00:17:22 - that's how this whole thing started in the first place, right? Router C
00:17:26 - heard about a route that it told router B about, and that's
00:17:30 - what caused router C to point to router B, and started the whole
00:17:32 - loop from the beginning.
00:17:34 - Now, you'll see when we get to some of the advanced WAN
00:17:37 - network diagrams, why this can cause some problems, especially
00:17:41 - in networks like frame relay. But for now, split horizon is definitely
00:17:46 - a good thing.
00:17:47 - Last, but not least, is the hold down timer; your best and worst
00:17:53 - friend when it comes to loop prevention.
00:17:57 - Hold down timers says, I will not believe any other updates
00:18:02 - about this route for x amount of time. Here's what I mean: When
00:18:07 - this network goes down, router C sends a triggered update
00:18:11 - saying, it's down. Router B receives that and says, well, that's, that's
00:18:16 - a big change. Let me immediately tell router A about it, and
00:18:19 - both of them will set a hold down timer. This is my little
00:18:23 - clock. Now, that hold down timer can vary depending on what protocol
00:18:27 - you're using, but what it does is it says, I will not accept
00:18:31 - any more updates about this for a certain amount of time. The
00:18:35 - reason that this is in place, is because of flapping interfaces.
00:18:41 - If you've never heard of a flapping interface before, what
00:18:44 - it is, is it's an interface that goes up and down, and up and down
00:18:47 - and up and down, and up and down; a million times a second. Or maybe
00:18:50 - not that many, but a lot. It's just constantly going up and down. It's
00:18:54 - caused by a bad network cable. It could be a bad connector, you
00:18:58 - know it's not plugged in quite all the way, but it's in most of the way.
00:19:01 - Or, it could just be a network interface card, a network card that's going
00:19:05 - bad. Now, the problem is, is we introduce this system of triggered
00:19:09 - updates. Well, if we've got that, that can destroy a network when
00:19:15 - combined with flapping updates. Because, or flapping interface,
00:19:18 - because when this goes up immediately router C says, hey, router B, it's up.
00:19:21 - Router B is like, hey, router A, it's up. Oh, nope, it's down. Down, down. And everybody's
00:19:26 - adding and removing that route 1,000 times a second from
00:19:29 - their routing table. All the routers on your network has their
00:19:32 - processor utilization shoot through the roof. So, what hold down
00:19:37 - timers do is say, ok, you told me it's down, I will not believe
00:19:41 - anything else about that. Now, that's great in the sense that
00:19:46 - you've prevented flapping interfaces, but it's bad in the sense
00:19:49 - that you've got this IT admin, right, who tripped over that
00:19:53 - cable, and he sees it. He goes, "Oh! I can't believe I did
00:19:57 - that." And, you know, plugs the cable back in. Well, immediately,
00:20:00 - router C is going to say, oh, ok, it's back up, but router B is not going
00:20:04 - to believe that. It's going to say, sorry, you've told me it's down,
00:20:08 - and I'm going to believe it's down for x more seconds. I think
00:20:13 - by default on a CISCO router, it's 180 seconds.
00:20:16 - 180 seconds! Can you believe that? That's a huge amount
00:20:21 - of downtime just because someone unplugged a cable
00:20:24 - and then tried to plug it back in. So, all of those five things
00:20:29 - are loop prevention mechanisms, and they are all drawbacks of
00:20:33 - using the RIP routing protocol.
00:20:36 - Now, let's change our focus to the link state routing protocols,
00:20:40 - which have no loops because they have a completely different
00:20:43 - system of handling routes and route updates. First off, link
00:20:48 - state routing protocols form neighbor relationships
00:20:52 - with each other, rather than just sending broadcast or
00:20:55 - multipackets, multicast packets to everyone. So, for example, I gave
00:20:59 - you the example of a distance vector routing protocol being
00:21:03 - like you walking in front of a group of people into a microphone
00:21:06 - and saying, hey, everyone, I know about these networks. Well, in
00:21:10 - a link state routing protocol, if it was the same situation, you
00:21:14 - would stand up on the stage and then you'd go, you know what,
00:21:17 - Let's, let's not do this. Let's walk down. And you would find
00:21:20 - the first person in that crowd that you would want to exchange
00:21:23 - route with, routes with and you would say, "Hi, my name is Jeremy. What's
00:21:26 - yours?" And they would say, "Hi, my name is, my name is Michael." And
00:21:30 - you would say, "Well, Michael", it's funny, I'm doing this odd system
00:21:35 - where I'm saying you, but it's really me. So, how about I flip it to me. I
00:21:38 - would say, "Hi Michael, I know about these routes," and Michael
00:21:42 - would say, "Great! Let me put them in my routing table. Jeremy, I know
00:21:45 - about these routes." And I'd say, "Great! Let me add those to my
00:21:49 - routing table." And then, we just look at each other, and I would
00:21:52 - say, "Hi Michael." And he'd look back and say, "Hi."
00:21:56 - And then I'd wait, and then I'd say, "Hi Michael." And he would say, "Hi." And
00:22:01 - we keep doing that all day long, "hi, hi, hi". That's actually the
00:22:05 - technical name of a protocol called the hello protocol. OSPF
00:22:10 - uses this protocol known as hello and it sends it much
00:22:14 - more frequently than RIP to make sure that the router is still
00:22:18 - online. Since I am just sending a single message to another
00:22:22 - router, I'm just saying "hello", I can send that much more often
00:22:26 - than RIP could when it was sending broadcasts and multicasts
00:22:29 - packets because it bothered the whole network. If you were to tune RIP down
00:22:32 - to send those more often, you would cause some major network
00:22:35 - problems. So,
00:22:38 - OSPF and other link state routing protocols use "hello".
00:22:42 - After the initial routing tables are exchanged, routers just
00:22:45 - send small, event-based updates. Meaning, with distance factor, we
00:22:50 - sent the whole routing table every 30 seconds. With link state routing
00:22:54 - protocols, I only send an update when I need to send an update.
00:22:59 - When something changes, I would say, "Hi Michael, 192.168.1.0
00:23:03 - just went down." And you go, "Oh that, that's great! I'll remove
00:23:07 - that from my table." And then I just look at him and go, "Hi Michael."
00:23:10 - Nothing more; I don't need to send any more updates, unlike
00:23:14 - RIP. So there are only two link state routing protocols
00:23:18 - that exist today. That is OSPF, that's what we're going to
00:23:21 - talk about here, and IS-IS, which is covered more in the
00:23:25 - CCNP curriculum.
00:23:28 - So, to wrap up this conceptual video on routing protocols, let's
00:23:32 - look at the advantages and disadvantages of using a link state
00:23:37 - protocol. Advantage: they are much faster to converge. They can
00:23:41 - find problems on the network and repair them much faster than
00:23:45 - distance vector. And the reason why, is they're saying hello
00:23:48 - more often so they can detect a failure much, much quicker. There's
00:23:53 - no routing loops because the routers have a complete road map of
00:23:57 - the network, which we'll see as we dig deeper into OSPF. So
00:24:01 - they don't, they don't ever get confused. They don't ever need
00:24:04 - loop prevention mechanisms. If you've got a road map, there's no chance
00:24:08 - that you're going in the wrong direction. Unless you don't know how to read
00:24:10 - a map.
00:24:11 - Finally, it forces your, you to design your network correctly.
00:24:16 - And I have that as both an advantage and, you can see, a disadvantage
00:24:22 - that's because link state routing protocols do require a solid
00:24:25 - network design. Meaning, if you've designed your network poorly and
00:24:29 - just assigned IP addresses where you wanted to assign them; put a network here,
00:24:33 - put in a network there. No real design in mind. Well, link state
00:24:37 - routing protocols are really not going to be an advantage to
00:24:40 - you, at that point, because they will cause worse problems on
00:24:43 - your network than if you had used distance vector.
00:24:46 - So, some of the disadvantages is: number one, they do consume
00:24:49 - more resources on your router. They'll use more processor and
00:24:53 - more memory utilization because they're simply more complex. They
00:24:57 - do require a solid network design, and I'm, I'm skating around that one
00:25:01 - for now, the deep definition of that, because as we look
00:25:04 - at OSPF specifically, I'll show you what a good design looks like. Finally,
00:25:09 - link state routing protocols have technical complexity. There's
00:25:14 - a lot to them and there's a lot to know about them if you want
00:25:17 - to use them efficiently in your network. So, there is a lot
00:25:21 - more education necessary to use link state protocols.
00:25:25 - We've now taken our first step back into the routing protocol
00:25:29 - world, since the ICND1 videos. As we've looked at distance
00:25:33 - vector versus link state. So, let me summarize, we looked at the two
00:25:38 - classes or umbrellas of routing protocols. All in all, distance
00:25:42 - vector, you can think of as easy to configure and featureless. Meaning,
00:25:46 - they don't have that many features, whereas link state you can
00:25:50 - think of as complex and require a lot of tuning, but they do
00:25:54 - have just about every feature you would expect from a routing
00:25:57 - protocol. So that's the difference. We then looked at the
00:26:01 - loop prevention mechanisms of distance vector, which are really
00:26:05 - a Band-aid for some of the problems introduced with the routing
00:26:08 - protocol that broadcasts in all directions everything it knows
00:26:11 - every 30 seconds. Finally, we looked at the advantages and
00:26:15 - disadvantages, and what the big difference is when using a link
00:26:20 - state routing protocol. I hope this has been informative for you
00:26:23 - and I'd like to thank you for viewing.