WEBVTT

0:00:08.340000 --> 0:00:12.660000
 So we're going to leave the world of
 EIGRP behind us and at this point

0:00:12.660000 --> 0:00:16.060000
 we're not going to move into another
 interior gateway routing protocol

0:00:16.060000 --> 0:00:21.540000
 which is open shortest path first
 otherwise known as OSPF.

0:00:21.540000 --> 0:00:27.020000
 So let's just do a quick
 introduction to OSPF.

0:00:27.020000 --> 0:00:32.340000
 First thing is is that OSPF is and
 always has been an open standard.

0:00:32.340000 --> 0:00:36.680000
 There actually is an RFC for it if
 you want to Google the RFC for OSPF

0:00:36.680000 --> 0:00:40.680000
 but it's not owned by any company, it's
 not proprietary, it's always been

0:00:40.680000 --> 0:00:42.360000
 an open standard.

0:00:42.360000 --> 0:00:44.900000
 This is actually very,
 very old as well.

0:00:44.900000 --> 0:00:49.000000
 It's been around for several
 decades at this point.

0:00:49.000000 --> 0:00:53.960000
 So OSPF uses something called the
 shortest path first algorithm.

0:00:53.960000 --> 0:00:55.160000
 What does that mean?

0:00:55.160000 --> 0:01:01.000000
 Well unlike EIGRP which is what we classify
 as an advanced distance vector

0:01:01.000000 --> 0:01:06.220000
 protocol, OSPF falls into a different
 category called a link state routing

0:01:06.220000 --> 0:01:10.220000
 protocol. One of the big differences
 between the two of them is that a

0:01:10.220000 --> 0:01:15.340000
 distance vector or advanced distance
 vector protocol all it knows about

0:01:15.340000 --> 0:01:19.160000
 as far as the topology is concerned
 is that it's self and has directly

0:01:19.160000 --> 0:01:21.700000
 connected neighbors and that's it.


0:01:21.700000 --> 0:01:24.940000
 So if you and I are running a distance
 vector routing protocol I will

0:01:24.940000 --> 0:01:28.580000
 tell you about all the routes I know
 about so you know that to get to

0:01:28.580000 --> 0:01:31.060000
 those routes you go through me.

0:01:31.060000 --> 0:01:35.080000
 So you know those routes live back here
 somewhere but you don't know what

0:01:35.080000 --> 0:01:37.400000
 the topology looks like back here.


0:01:37.400000 --> 0:01:41.620000
 You have no idea what is connected
 to what, what the links look like,

0:01:41.620000 --> 0:01:46.040000
 no idea. Whereas in a link state routing
 protocol you actually know the

0:01:46.040000 --> 0:01:49.840000
 entire topology not just you and your
 directly connected neighbors but

0:01:49.840000 --> 0:01:54.920000
 everything. You can form like a map
 in your own mind as to where all the

0:01:54.920000 --> 0:01:58.860000
 routers are, where the subnets sit on
 each one of those routers and you

0:01:58.860000 --> 0:02:00.540000
 can piece it together.

0:02:00.540000 --> 0:02:05.200000
 So the shortest path first algorithm
 takes place after you've built that

0:02:05.200000 --> 0:02:08.540000
 map in your mind it's sort of like
 spanning tree where you're like the

0:02:08.540000 --> 0:02:14.260000
 root of the tree and everything path
 first algorithm you build this tree

0:02:14.260000 --> 0:02:17.520000
 and then you figure out from your perspective
 is you at the center of

0:02:17.520000 --> 0:02:21.780000
 the universe what the shortest path
 is to any given destination.

0:02:21.780000 --> 0:02:24.420000
 That's the SPF algorithm.

0:02:24.420000 --> 0:02:31.120000
 Like EIGRP OSPF utilizes Hello Packets
 to dynamically discover and build

0:02:31.120000 --> 0:02:34.040000
 OSPF neighbor relationships.

0:02:34.040000 --> 0:02:38.360000
 In the case of OSPF we actually call
 them adjacencies OSPF adjacencies.

0:02:38.360000 --> 0:02:42.560000
 So I'll use those terms interchangeably,
 neighborship adjacency meaning

0:02:42.560000 --> 0:02:44.380000
 basically the same thing.

0:02:44.380000 --> 0:02:51.100000
 Like EIGRP where the vast majority of
 EIGRP packets went to a multicast

0:02:51.100000 --> 0:02:55.720000
 address most OSPF packets also
 go to multicast addresses.

0:02:55.720000 --> 0:03:00.400000
 Matter of fact OSPF has two addresses
 that are reserved just for itself

0:03:00.400000 --> 0:03:05.900000
 224 0 0 5 and 224 0 0 6.

0:03:05.900000 --> 0:03:11.560000
 OSPF Hello Packets always
 go to 224 0 0 5.

0:03:11.560000 --> 0:03:15.700000
 The other packets that we're going to
 talk about with OSPF may go to dot

0:03:15.700000 --> 0:03:17.780000
 5 or they may go to dot 6.

0:03:17.780000 --> 0:03:20.200000
 When we get a little bit further into
 this and I talk about this concept

0:03:20.200000 --> 0:03:25.200000
 of something called designated routers
 and backup designated routers the

0:03:25.200000 --> 0:03:28.160000
 idea between these two different addresses
 will make a little bit more

0:03:28.160000 --> 0:03:36.100000
 sense. OSPF also works based on an area
 hierarchy minimizing LSA flooding.

0:03:36.100000 --> 0:03:37.640000
 What does that talk about?

0:03:37.640000 --> 0:03:40.660000
 We haven't talked about
 areas or LSA's yet.

0:03:40.660000 --> 0:03:43.100000
 Let's hold off on that bullet
 for just a minute.

0:03:43.100000 --> 0:03:46.220000
 The very next slide is going to go into
 a little bit more detail of what

0:03:46.220000 --> 0:03:49.040000
 that acronym LSA means.

0:03:49.040000 --> 0:03:53.040000
 And once I talk about LSA's then I'll be
 able to give you a better understanding

0:03:53.040000 --> 0:04:00.440000
 of what an OSPF area is and why you
 might want more than one area.

0:04:00.440000 --> 0:04:03.700000
 And like pretty much all the routing
 protocols that utilize neighbors

0:04:03.700000 --> 0:04:09.200000
 OSPF has the optional capability of
 supporting authentication both clear

0:04:09.200000 --> 0:04:12.820000
 text and MD 5 authentication.

0:04:12.820000 --> 0:04:18.400000
 Okay, so the real heart and soul of
 OSPF are these things called link

0:04:18.400000 --> 0:04:20.740000
 state advertisements.

0:04:20.740000 --> 0:04:25.100000
 What is that? Okay, first of all remember
 that in OSPF we call this a

0:04:25.100000 --> 0:04:27.780000
 link state routing protocol.

0:04:27.780000 --> 0:04:31.080000
 In other words, if I'm a router and
 this is what I would normally call

0:04:31.080000 --> 0:04:35.460000
 an interface like interface fast ethernet
 zero zero or interface serial

0:04:35.460000 --> 0:04:41.000000
 zero zero OSPF calls that a link
 not an interface a link.

0:04:41.000000 --> 0:04:45.800000
 Technically speaking everything that
 has a subnet would be considered

0:04:45.800000 --> 0:04:50.320000
 a link. So if I have a sub
 interface that's a link.

0:04:50.320000 --> 0:04:54.840000
 If I have a physical interface that
 has two or three subnets on it that

0:04:54.840000 --> 0:04:58.540000
 would be categorized as two or three
 separate links even though it's on

0:04:58.540000 --> 0:05:00.800000
 one physical interface.

0:05:00.800000 --> 0:05:02.960000
 So it's a link we're
 talking about.

0:05:02.960000 --> 0:05:06.800000
 The next thing is as an OSPF router
 if I'm directly connected to this

0:05:06.800000 --> 0:05:11.120000
 interface right here let's say it's fast
 ethernet one then it's my responsibility

0:05:11.120000 --> 0:05:16.000000
 to create a special type of a packet
 called a link state update packet

0:05:16.000000 --> 0:05:19.660000
 and LSU. I'll go ahead
 and write that on here.

0:05:19.660000 --> 0:05:27.660000
 LSU which stands for link state update
 and within that link state update

0:05:27.660000 --> 0:05:34.860000
 packet I will pack in there one or
 more link state advertisements.

0:05:34.860000 --> 0:05:39.500000
 So an LSU packet think of that as being
 like in EIGRP terminology and

0:05:39.500000 --> 0:05:44.680000
 EIGRP update. An EIGRP update contains
 one or more networks inside of

0:05:44.680000 --> 0:05:48.140000
 it one or more prefixes that
 that update is carrying.

0:05:48.140000 --> 0:05:54.320000
 Similarly an OSPF link state update
 contains one or more LSA's inside

0:05:54.320000 --> 0:05:57.140000
 of it. Now what exactly is an LSA?


0:05:57.140000 --> 0:06:03.840000
 Okay so let's say I need to advertise
 information about this link to all

0:06:03.840000 --> 0:06:05.820000
 the other OSPF routers.

0:06:05.820000 --> 0:06:10.580000
 Well I'm going to create an LSU I'm going
 to create a data structure called

0:06:10.580000 --> 0:06:14.140000
 a link state advertisement and that
 link state advertisement is going

0:06:14.140000 --> 0:06:18.120000
 to give you everything you need
 to know about this link.

0:06:18.120000 --> 0:06:20.560000
 Now before I go into the details of
 that here's the main thing I want

0:06:20.560000 --> 0:06:21.900000
 you to think about.

0:06:21.900000 --> 0:06:28.920000
 Number one the overall goal or objective
 of OSPF is to build this topology

0:06:28.920000 --> 0:06:30.280000
 map in its mind.

0:06:30.280000 --> 0:06:35.540000
 It wants to build this tree of what
 the entire topology looks like who's

0:06:35.540000 --> 0:06:40.320000
 connected to who who they're connected
 to that's its overall objective.

0:06:40.320000 --> 0:06:42.680000
 So imagine just close your eyes for
 a second imagine a whole bunch of

0:06:42.680000 --> 0:06:46.300000
 circles and you're starting to draw lines
 between those circles and you're

0:06:46.300000 --> 0:06:49.760000
 building a tree like structure interconnecting
 these circles and the circles

0:06:49.760000 --> 0:06:53.240000
 are representing routers and the lines
 are representing links connecting

0:06:53.240000 --> 0:06:57.980000
 them. So you're building that and then
 as you build that imagine you start

0:06:57.980000 --> 0:07:03.620000
 sort of writing in the various subnets
 where they live on top of the links.

0:07:03.620000 --> 0:07:06.680000
 That's sort of your secondary objective
 is to figure out where the subnets

0:07:06.680000 --> 0:07:10.460000
 are but your primary objective
 is to build this tree.

0:07:10.460000 --> 0:07:13.260000
 Build this SPF tree.

0:07:13.260000 --> 0:07:17.940000
 Okay so the link state advertisement
 that's its main job is to help you

0:07:17.940000 --> 0:07:19.600000
 figure out this topology.

0:07:19.600000 --> 0:07:23.840000
 So as such when I create this link
 state advertisement describing fast

0:07:23.840000 --> 0:07:28.900000
 ethernet one I'm going to tell you a
 whole lot more than just what subnet

0:07:28.900000 --> 0:07:32.680000
 lives here. You see if I was a distance
 vector routing protocol that's

0:07:32.680000 --> 0:07:35.980000
 all you care about is what subnets I'm
 connected to you could care less

0:07:35.980000 --> 0:07:40.200000
 about anything else in a link state
 advertisement I'm going to say this

0:07:40.200000 --> 0:07:42.740000
 is a broadcast link.

0:07:42.740000 --> 0:07:45.360000
 This link has a certain bandwidth.


0:07:45.360000 --> 0:07:49.320000
 Let me tell you about all the neighbors
 I'm connected to on this link.

0:07:49.320000 --> 0:07:53.160000
 Oh let me also tell you about the
 subnet on this link as well.

0:07:53.160000 --> 0:07:56.620000
 I'll actually tell you all sorts of
 descriptive information about this

0:07:56.620000 --> 0:08:02.200000
 link in my LSA. And now here's
 another big difference.

0:08:02.200000 --> 0:08:05.480000
 In the world of EIGRP let's
 do some contrasting here.

0:08:05.480000 --> 0:08:09.460000
 Let's say that you and I are two routers
 connected via EIGRP and behind

0:08:09.460000 --> 0:08:14.300000
 you you have another EIGRP neighbor relationship
 with someone behind you.

0:08:14.300000 --> 0:08:16.740000
 So I send you an EIGRP update.

0:08:16.740000 --> 0:08:20.700000
 In my EIGRP update I say hey my name
 is router Keith that's my router

0:08:20.700000 --> 0:08:25.500000
 ID and here's a bunch of networks
 you can get through me to reach.

0:08:25.500000 --> 0:08:29.680000
 Okay now what do you do when you turn
 around and send an EIGRP update

0:08:29.680000 --> 0:08:34.860000
 to your neighbor do you tell your neighbor
 about me about my router ID

0:08:34.860000 --> 0:08:38.320000
 and that these networks
 you learned from me?

0:08:38.320000 --> 0:08:41.940000
 No you don't. Let's say
 your name is Sally.

0:08:41.940000 --> 0:08:44.760000
 Well when you turn around and you send
 an EIGRP update to your neighbor

0:08:44.760000 --> 0:08:49.120000
 all you say is you copy those networks
 I gave you and you say hey my name

0:08:49.120000 --> 0:08:53.560000
 is Sally neighbor and I'm Sally and
 here's the networks I know about.

0:08:53.560000 --> 0:08:58.100000
 So your neighbor knows about the networks
 are reachable through you but

0:08:58.100000 --> 0:09:01.680000
 your neighbor has no idea that I'm
 the one who actually told you about

0:09:01.680000 --> 0:09:06.920000
 those networks and similarly you have
 no idea about who told me about

0:09:06.920000 --> 0:09:11.380000
 those networks. Well an OSPF
 is just the opposite.

0:09:11.380000 --> 0:09:15.020000
 If I created link state advertisement
 and I package one or more of those

0:09:15.020000 --> 0:09:21.620000
 inside a LSU link state update packet
 and I multi-cast it to you you take

0:09:21.620000 --> 0:09:25.800000
 that link state advertisement and stick
 in a database you store it and

0:09:25.800000 --> 0:09:28.200000
 then you keep right
 on flooding it.

0:09:28.200000 --> 0:09:30.220000
 You don't change anything
 about it.

0:09:30.220000 --> 0:09:35.520000
 So even if there's nine or ten routers
 behind you they will all know about

0:09:35.520000 --> 0:09:38.060000
 me. They will know about my name.

0:09:38.060000 --> 0:09:41.060000
 They will know that I'm the one that
 created this LSA that I'm the one

0:09:41.060000 --> 0:09:42.540000
 who owns this link.

0:09:42.540000 --> 0:09:45.620000
 So they will be able to in their mind
 as they're putting their jigsaw

0:09:45.620000 --> 0:09:50.780000
 puzzle together they will know that
 I exist and where I sit and they'll

0:09:50.780000 --> 0:09:53.560000
 know that I'm connected to you and that
 you and I are neighbors and I'm

0:09:53.560000 --> 0:09:55.000000
 neighbors with some
 other people here.

0:09:55.000000 --> 0:09:59.620000
 So the LSA gets flooded to all these
 other routers and that's how they

0:09:59.620000 --> 0:10:06.340000
 build this topology diagram.

0:10:06.340000 --> 0:10:10.900000
 So an LSA is carried within a link state
 update packet and different types

0:10:10.900000 --> 0:10:15.360000
 of LSA carry different data so this
 is another important ingredient.

0:10:15.360000 --> 0:10:20.000000
 So at the CCNA level they don't really
 expect you to know the details

0:10:20.000000 --> 0:10:25.080000
 of what exactly an LSA is and that there's
 different kinds and what the

0:10:25.080000 --> 0:10:29.580000
 names of those different kinds are and
 what different purposes they serve

0:10:29.580000 --> 0:10:34.940000
 but you sort of get the idea here that
 there are different kinds of LSA's

0:10:34.940000 --> 0:10:41.460000
 and they do different things depending
 on where a router sits in the topology.

0:10:41.460000 --> 0:10:47.200000
 And LSA is aged out after one hour
 and are refreshed every 30 minutes.

0:10:47.200000 --> 0:10:51.760000
 So what that means is that this LSA
 I create describing this link when

0:10:51.760000 --> 0:10:55.120000
 I flooded it to you and then it got
 flooded to all the other people out

0:10:55.120000 --> 0:11:01.420000
 there well you and I are exchanging
 hello packets just like an EIGRP.

0:11:01.420000 --> 0:11:04.660000
 You and I are exchanging hello packets
 and that's how you know I still

0:11:04.660000 --> 0:11:08.240000
 exist that I'm still a valid neighbor
 because every few seconds you're

0:11:08.240000 --> 0:11:09.980000
 getting a hello packet from me.

0:11:09.980000 --> 0:11:14.640000
 Now if we were doing EIGRP once I sent
 you an EIGRP update with some networks

0:11:14.640000 --> 0:11:18.700000
 in it you might never see an
 update from me ever again.

0:11:18.700000 --> 0:11:22.360000
 I might never tell you about those networks
 for an hour a week six months

0:11:22.360000 --> 0:11:26.600000
 hey as long as you and I are still exchanging
 hello packets you say hey

0:11:26.600000 --> 0:11:30.560000
 Keith still exists so that routing
 update that Keith sent me about the

0:11:30.560000 --> 0:11:36.280000
 75-75 network three months ago must still
 be there because he hasn't told

0:11:36.280000 --> 0:11:37.880000
 me it's not there.

0:11:37.880000 --> 0:11:47.000000
 OSPF takes a little bit of time to
 get to all the people behind you.

0:11:47.000000 --> 0:11:50.340000
 So yeah it might be well and good for
 you and I do exchange hello and

0:11:50.340000 --> 0:11:54.800000
 you know that I'm here but
 how do they know I'm here?

0:11:54.800000 --> 0:11:58.400000
 I'm not sending hello packets to the
 people behind you I don't have a

0:11:58.400000 --> 0:12:02.880000
 neighbor relationship with them I have
 it with you so at some point in

0:12:02.880000 --> 0:12:06.680000
 time the people behind you might be
 saying oh it's been a while since

0:12:06.680000 --> 0:12:11.220000
 we got some LSA from some guy out there
 named Keith is he still alive

0:12:11.220000 --> 0:12:17.480000
 is he actually still reachable so in
 OSPF we say well when an LSA goes

0:12:17.480000 --> 0:12:21.740000
 out there's a timer associated with it
 if you don't get that LSA refreshed

0:12:21.740000 --> 0:12:26.560000
 one hour later it purges from your
 database you delete it you say okay

0:12:26.560000 --> 0:12:30.780000
 it's been an hour haven't heard from
 that guy he must be gone I'll delete

0:12:30.780000 --> 0:12:35.740000
 the LSA so as the owner of the LSA
 is the person who actually created

0:12:35.740000 --> 0:12:42.060000
 it it's my responsibility every 30 minutes
 to refresh that LSA so I will

0:12:42.060000 --> 0:12:48.140000
 reflet it every 30 minutes okay so
 let's now let's go back to what we

0:12:48.140000 --> 0:12:53.040000
 were just talking about this slide right
 here and I want to pay some special

0:12:53.040000 --> 0:12:59.480000
 attention to works on an area hierarchy
 to minimize LSA flooding and this

0:12:59.480000 --> 0:13:04.860000
 will hopefully make a little bit more
 sense now so let's look at this

0:13:04.860000 --> 0:13:11.320000
 topology diagram right here so the
 one there's a lot of rules that go

0:13:11.320000 --> 0:13:15.540000
 along with OSPF and this is why some people
 don't like OSPF because there's

0:13:15.540000 --> 0:13:19.180000
 a lot of rules to remember about what
 it can do and what it can't do and

0:13:19.180000 --> 0:13:22.960000
 what one router's job is versus what
 another router's job is so I'm going

0:13:22.960000 --> 0:13:31.440000
 to give you one of the rules right
 now so one of the rules states that

0:13:31.440000 --> 0:13:38.420000
 if you ever receive an LSA describing
 that something in the tree has changed

0:13:38.420000 --> 0:13:43.200000
 right so imagine for a moment you're an
 OSPF router you've had some neighbor

0:13:43.200000 --> 0:13:46.700000
 relationships going now for several
 hours or several days everything's

0:13:46.700000 --> 0:13:49.960000
 been nice and steady you're exchanging
 hello packets every few seconds

0:13:49.960000 --> 0:13:53.980000
 and you've built this nice tree you know
 where everybody is and everybody's

0:13:53.980000 --> 0:13:56.580000
 sitting and you've got this nice tree
 in your mind now all of a sudden

0:13:56.580000 --> 0:14:01.100000
 here comes in an LSA from some other
 router out there who says there's

0:14:01.100000 --> 0:14:06.120000
 a new link in the tree or who says an
 existing link has changed it's gone

0:14:06.120000 --> 0:14:10.820000
 down the rule of OSPF states that when
 that happens you have to take that

0:14:10.820000 --> 0:14:15.100000
 nice tree and rip it apart it's like
 taking your jigsaw puzzle and ripping

0:14:15.100000 --> 0:14:19.660000
 it apart and then build it all back
 together again accounting for that

0:14:19.660000 --> 0:14:26.620000
 new LSA that's an important point here's
 why this process of building

0:14:26.620000 --> 0:14:31.820000
 a tree is very CPU intensive it takes
 the brain of the router a lot of

0:14:31.820000 --> 0:14:36.940000
 power to put all these LSA's together
 to see who connects with who and

0:14:36.940000 --> 0:14:41.440000
 to eventually build this map of the
 topology in its mind so every time

0:14:41.440000 --> 0:14:45.340000
 an LSA comes in and says oh man I got
 to rip apart my tree and now rebuild

0:14:45.340000 --> 0:14:49.500000
 it with this one new jigsaw puzzle
 or if an LSA comes in and says this

0:14:49.500000 --> 0:14:53.920000
 branch is gone this link is gone oh
 I got to rip apart my tree and now

0:14:53.920000 --> 0:14:58.420000
 rebuild it and exclude that branch that
 takes a lot of processing power

0:14:58.420000 --> 0:15:02.640000
 for the router to do that so imagine
 for a moment that we're looking at

0:15:02.640000 --> 0:15:07.940000
 this topology right here and let's
 say that there's some link in this

0:15:07.940000 --> 0:15:14.000000
 topology that is not very stable maybe
 it's this link right here okay

0:15:14.000000 --> 0:15:19.260000
 so that link is flapping for whatever
 reason it's going down and it's

0:15:19.260000 --> 0:15:25.060000
 coming up going down coming up well
 every time that happens these two

0:15:25.060000 --> 0:15:29.860000
 routers here router four and switch
 three are creating new LSA's about

0:15:29.860000 --> 0:15:34.960000
 a link down condition or a new link coming
 up condition which means those

0:15:34.960000 --> 0:15:40.280000
 LSA's are being flooded all over the
 place and so every time that happens

0:15:40.280000 --> 0:15:45.780000
 all these devices are having to tear
 down their tree rebuild it tear down

0:15:45.780000 --> 0:15:54.440000
 their tree rebuild it so in OSPF when
 you configure OSPF you're going

0:15:54.440000 --> 0:16:02.560000
 to start by doing router OSPF and then
 you're going to give it some process

0:16:02.560000 --> 0:16:06.040000
 ID and I'll talk more about the process
 ID coming up in another video

0:16:06.040000 --> 0:16:12.860000
 and then like EIGRP you're going to
 have some sort of network command

0:16:12.860000 --> 0:16:16.920000
 and let's just take switch three and exam
 as an example I'll have a network

0:16:16.920000 --> 0:16:22.200000
 command to cover this network this
 network and this network but unlike

0:16:22.200000 --> 0:16:31.280000
 EIGRP every link in OSPF is considered
 to be part of an area let's start

0:16:31.280000 --> 0:16:34.540000
 real quickly and just say that well
 we're going to have all these links

0:16:34.540000 --> 0:16:42.300000
 in this entire topology being one giant
 area now an area like an autonomous

0:16:42.300000 --> 0:16:48.600000
 system number has a number as a value
 if I'm having my entire topology

0:16:48.600000 --> 0:16:53.720000
 in just one area that area can be whatever
 I want well maybe we'll say

0:16:53.720000 --> 0:17:02.760000
 area eight so in switch three here
 I'd have network two three two and

0:17:02.760000 --> 0:17:07.500000
 area eight network three three three
 and area eight network three four

0:17:07.500000 --> 0:17:11.700000
 three and area eight and that would
 be enabling OSPF on these links so

0:17:11.700000 --> 0:17:16.400000
 we could start speaking about it so you
 could do that you could have every

0:17:16.400000 --> 0:17:21.160000
 link and your entire topology in one giant
 area now Cisco recommends there's

0:17:21.160000 --> 0:17:23.860000
 actually design recommendation that
 says you should have no more than

0:17:23.860000 --> 0:17:30.420000
 about 40 or 50 routers total in any
 one given area so right now this is

0:17:30.420000 --> 0:17:33.700000
 easy right so I've got one two three
 four five six seven I've got about

0:17:33.700000 --> 0:17:40.820000
 seven devices no problem here's the problem
 though or the potential problem

0:17:40.820000 --> 0:17:48.100000
 within an area everybody has to know
 the entire topology of that area

0:17:48.100000 --> 0:17:52.600000
 and everybody's database that's storing
 the link state advertisements

0:17:52.600000 --> 0:17:59.160000
 has to be the same and this is another
 rule of OSPF if you and I are OSPF

0:17:59.160000 --> 0:18:04.520000
 neighbors and we are what's called fully
 adjacent what that means is that

0:18:04.520000 --> 0:18:09.640000
 all the LSA's I have in my database
 you have to have them and whatever

0:18:09.640000 --> 0:18:13.760000
 you have in your database I have to
 have them our databases have to be

0:18:13.760000 --> 0:18:18.120000
 synchronized they have to look identical
 so in this particular topology

0:18:18.120000 --> 0:18:22.180000
 here all seven of these devices would
 have their own LSA's that they created

0:18:22.180000 --> 0:18:27.640000
 as well as the LSA's of all the other
 devices in here so even in the small

0:18:27.640000 --> 0:18:31.720000
 topologies we see here that'd be a
 fair amount of LSA's we've got one

0:18:31.720000 --> 0:18:37.600000
 two three four five six seven we've got
 about eight links so we have several

0:18:37.600000 --> 0:18:42.880000
 LSA's so all seven of these devices if
 you looked in any one of them their

0:18:42.880000 --> 0:18:45.980000
 link state database would be identical
 they'd all have exactly the same

0:18:45.980000 --> 0:18:51.900000
 LSA's and this is the reason why if
 a new LSA comes in everybody's got

0:18:51.900000 --> 0:18:55.080000
 to add it to their database because
 they're synchronized and they have

0:18:55.080000 --> 0:18:59.200000
 to tear apart their tree and rebuild
 it and then if another LSA comes

0:18:59.200000 --> 0:19:05.100000
 in synchronize it to their database tear
 apart their tree rebuild it well

0:19:05.100000 --> 0:19:09.700000
 if my topology is fairly stable I don't
 really have to worry about that

0:19:09.700000 --> 0:19:13.400000
 it's that environment I talked about
 earlier I says hey a link's flapping

0:19:13.400000 --> 0:19:17.120000
 and now it's causing the CPUs of my
 routers to get pretty high because

0:19:17.120000 --> 0:19:20.220000
 they have to keep having to tear apart
 their topology and rebuild it over

0:19:20.220000 --> 0:19:26.480000
 and over again so OSPF the designer
 said we're gonna give you a way to

0:19:26.480000 --> 0:19:33.060000
 avoid unnecessary CPU computations in
 that and in that type of topology

0:19:33.060000 --> 0:19:38.280000
 here's what we're going to do what you
 can do is you can take your topology

0:19:38.280000 --> 0:19:43.860000
 and instead of putting all of it in one
 area you can divide up into multiple

0:19:43.860000 --> 0:19:51.060000
 areas so for example I could say that
 all of these links here are in area

0:19:51.060000 --> 0:20:00.260000
 eight and all of these links are in
 another area now the rule of OSPF

0:20:00.260000 --> 0:20:05.600000
 here's another rule third rule the rule
 of OSPF says if you have two or

0:20:05.600000 --> 0:20:14.860000
 more areas one of them has to be the
 backbone area the backbone area and

0:20:14.860000 --> 0:20:22.000000
 that has a special number reserved
 area zero now I could have made the

0:20:22.000000 --> 0:20:26.300000
 green area area zero and I could have
 made the red area area eight doesn't

0:20:26.300000 --> 0:20:32.620000
 really matter as long as one of these
 two areas was area zero okay so

0:20:32.620000 --> 0:20:37.860000
 now why might I divide this up into
 two areas here's why a router only

0:20:37.860000 --> 0:20:42.980000
 has to know the full map of what the
 topology looks like of the areas

0:20:42.980000 --> 0:20:47.640000
 that it is connected to if there's other
 areas out there that this router

0:20:47.640000 --> 0:20:51.980000
 is not connected to he has no links
 in those other areas he doesn't know

0:20:51.980000 --> 0:20:55.140000
 what those areas look like they're
 just like a cloud to him he'll know

0:20:55.140000 --> 0:20:58.480000
 what subnets live there what networks
 are out there but he won't know

0:20:58.480000 --> 0:21:04.620000
 what that topology looks like so for
 example router four over here router

0:21:04.620000 --> 0:21:11.600000
 four knows exactly what's inside this
 green area he knows all of this

0:21:11.600000 --> 0:21:15.380000
 but as far as what's in this red area
 just think of that as being like

0:21:15.380000 --> 0:21:19.840000
 a cloud router four will know that
 the two two two network exists that

0:21:19.840000 --> 0:21:23.160000
 the one two one network exists that
 the one one one network exists and

0:21:23.160000 --> 0:21:27.580000
 that they're all reachable via switch
 two but he won't have any idea what

0:21:27.580000 --> 0:21:31.640000
 the topology looks like behind switch
 two that's sort of like a distance

0:21:31.640000 --> 0:21:37.900000
 vector characteristic actually now
 switch to because he's connected to

0:21:37.900000 --> 0:21:42.040000
 both areas he's actually going to have
 two trees that he has to build

0:21:42.040000 --> 0:21:46.760000
 he has to build a a shores path tree
 for area eight and a shores path

0:21:46.760000 --> 0:21:51.640000
 tree for area zero and he's actually going
 to have two databases one database

0:21:51.640000 --> 0:21:56.100000
 for all the lsa's he's collected in
 area eight and another database for

0:21:56.100000 --> 0:21:59.880000
 all the lsa's he's collected in area
 zero so he's going to have to work

0:21:59.880000 --> 0:22:04.940000
 a little bit harder as a matter of fact
 when a router such as switch to

0:22:04.940000 --> 0:22:08.600000
 i'm calling him switch to but we're
 just treating him as a router when

0:22:08.600000 --> 0:22:14.840000
 a router connects to area zero and
 one other area when he's sitting as

0:22:14.840000 --> 0:22:19.420000
 a border between the backbone area and
 a non-back bone area he actually

0:22:19.420000 --> 0:22:28.520000
 goes by the name of an area border router
 and abr so why is that important

0:22:28.520000 --> 0:22:35.600000
 well let's go back to this link right
 here so when switch three came online

0:22:35.600000 --> 0:22:43.900000
 and i activated ospf on this link he
 created an lsa that described that

0:22:43.900000 --> 0:22:52.960000
 link and that lsa got flooded to everybody
 now that lsa is i'll just give

0:22:52.960000 --> 0:22:56.960000
 you a preview of it i said there's different
 kinds of lsa's this lsa here

0:22:56.960000 --> 0:23:08.020000
 in green i'm describing is called a type
 one type one router lsa the primary

0:23:08.020000 --> 0:23:15.060000
 objective of a type one router lsa
 is to help construct this tree help

0:23:15.060000 --> 0:23:20.420000
 build this spf tree a type one router
 lsa will get flooded within an area

0:23:20.420000 --> 0:23:26.980000
 but it has to stop at the area border
 router in other words all these

0:23:26.980000 --> 0:23:31.560000
 devices in here in the green area area
 eight they need to see each other's

0:23:31.560000 --> 0:23:35.480000
 type one router lsa's that's what helps
 them build this topology but those

0:23:35.480000 --> 0:23:39.900000
 type one router lsa's will stop at
 switch two they will not be allowed

0:23:39.900000 --> 0:23:45.260000
 to go into the backbone area into area
 zero similarly there's going to

0:23:45.260000 --> 0:23:49.720000
 be type one router lsa's that are generated
 here in area zero so all the

0:23:49.720000 --> 0:23:52.520000
 devices in area zero can know what
 their topology looks like and they

0:23:52.520000 --> 0:23:57.180000
 will stop at the area border router now
 you might be thinking wait a second

0:23:57.180000 --> 0:24:02.580000
 if those type one router lsa stop at
 the abr and go no further then how

0:24:02.580000 --> 0:24:07.160000
 is area zero ever going to learn about
 the routes over here in area eight

0:24:07.160000 --> 0:24:11.780000
 and how is area eight ever going to
 learn about the routes over here in

0:24:11.780000 --> 0:24:15.360000
 area zero that's one of the special
 things the area border router does

0:24:15.360000 --> 0:24:21.460000
 he says well i'm not allowed to flood
 router lsa's from one area into

0:24:21.460000 --> 0:24:26.580000
 another area i can't do that so hmm
 what should i do what on i create

0:24:26.580000 --> 0:24:31.500000
 a different kind of an lsa an lsa whose
 purpose is not to create a tree

0:24:31.500000 --> 0:24:36.080000
 not to create any kind of topology information
 and lsa whose sole purpose

0:24:36.080000 --> 0:24:40.540000
 is just to describe networks just describe
 what subnets live in various

0:24:40.540000 --> 0:24:44.960000
 areas and that's what the abr will do
 he will actually create a special

0:24:44.960000 --> 0:24:50.040000
 type of an lsa actually several of
 them that will go out this way and

0:24:50.040000 --> 0:24:54.740000
 these lsa's will say hey i'm your abr
 i'm switch to and you can use me

0:24:54.740000 --> 0:25:00.020000
 to get to the 232 network or the 323
 or the 333 or the 343 you can get

0:25:00.020000 --> 0:25:03.620000
 there through me i'm not telling you
 what the topology looks like but

0:25:03.620000 --> 0:25:06.760000
 with this special kind of edel say i'm
 just saying there's another area

0:25:06.760000 --> 0:25:11.400000
 you guys don't know about in area zero
 and you can reach those subnets

0:25:11.400000 --> 0:25:17.160000
 through me so that's why the abr is
 very special once he realizes oh hey

0:25:17.160000 --> 0:25:23.320000
 i connect to the backbone area and one
 or more non backbone areas he says

0:25:23.320000 --> 0:25:28.560000
 okay router lsa's stop with me and
 i have to generate a different kind

0:25:28.560000 --> 0:25:32.340000
 of an lsa so that these two areas can
 learn about the routes that live

0:25:32.340000 --> 0:25:38.080000
 there so that's why we say that ospf
 works on an area hierarchy an area

0:25:38.080000 --> 0:25:43.300000
 hierarchy so just to once again put
 that in a perspective if i have one

0:25:43.300000 --> 0:25:50.900000
 area that can be whatever i want but
 if i have two or more areas like

0:25:50.900000 --> 0:25:58.560000
 this and maybe like this then the area
 in the middle is going to be my

0:25:58.560000 --> 0:26:05.560000
 backbone area which is area zero the
 other areas can be whatever i want

0:26:05.560000 --> 0:26:16.420000
 them to be maybe this is area 20 and
 maybe over here this is area 77 but

0:26:16.420000 --> 0:26:23.280000
 the main idea is that everything has
 to flow through the backbone so even

0:26:23.280000 --> 0:26:28.080000
 theoretically let's say i had a router
 here who is like this say this

0:26:28.080000 --> 0:26:34.260000
 is router x router x has a link in
 area 20 he has a link in area 77 is

0:26:34.260000 --> 0:26:40.740000
 he an abr no he is not because he is
 not connected to area zero so as

0:26:40.740000 --> 0:26:45.280000
 such router x does know what the topology
 looks like in area 20 and what

0:26:45.280000 --> 0:26:49.800000
 the topology looks like in area 77 but
 he is not allowed to send routes

0:26:49.800000 --> 0:26:54.860000
 from area 77 into area 20 and vice versa
 anything he learns about in area

0:26:54.860000 --> 0:26:59.220000
 20 he is not allowed to send those routes
 into area 77 because he is not

0:26:59.220000 --> 0:27:04.240000
 an abr so these other routers right
 here that are between the backbone

0:27:04.240000 --> 0:27:10.780000
 area and the non backbone areas these
 are my abr's so routes will always

0:27:10.780000 --> 0:27:19.240000
 be learned from the backbone and then
 into non backbone areas that's the

