WEBVTT

1
00:00.000 --> 00:03.000
So let's begin with some fundamentals we're getting OSPF.

2
00:03.000 --> 00:05.000
We'll get into the areas here in just a moment.

3
00:05.000 --> 00:08.000
So let's imagine we have these two routers right here

4
00:08.000 --> 00:12.000
and they are both configured for OSPF process ID1, for example,

5
00:12.000 --> 00:13.000
on each of them.

6
00:13.000 --> 00:16.000
And all the interfaces have been associated with area 0.

7
00:16.000 --> 00:20.000
Now, that also is one of the concepts I didn't really grasp on to

8
00:20.000 --> 00:23.000
when I was first learning OSPF like 30 years ago.

9
00:23.000 --> 00:26.000
I thought, "Oh, a router belongs to an OSPF area."

10
00:26.000 --> 00:29.000
But in reality, when we configure OSPF on a router,

11
00:29.000 --> 00:32.000
what we're doing is we're assigning interfaces to the areas.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

11
00:29.000 --> 00:32.000
what we're doing is we're assigning interfaces to the areas.

12
00:32.000 --> 00:36.000
And that's what makes that router, for example, belong to area 0.

13
00:36.000 --> 00:39.000
Every single interface on that router that's running OSPF

14
00:39.000 --> 00:43.000
is enabled in associated with area 0, also known as the backbone area.

15
00:43.000 --> 00:46.000
That's what makes this router an OSPF backbone router

16
00:46.000 --> 00:49.000
because all of its interfaces are in the backbone in area 0.

17
00:49.000 --> 00:52.000
And we'll break that into larger areas here in just a moment.

18
00:52.000 --> 00:55.000
But let's imagine that router 1 and router 2 have both been configured.

19
00:55.000 --> 00:59.000
All of their interfaces have been enabled for OSPF and associated with area 0.

20
00:59.000 --> 01:02.000
And they want to go ahead and start neighboring up.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

20
00:59.000 --> 01:02.000
And they want to go ahead and start neighboring up.

21
01:02.000 --> 01:04.000
They want to build an OSPF full of Jocency.

22
01:04.000 --> 01:07.000
Now, what is going to stop that from happening?

23
01:07.000 --> 01:09.000
There's a lot that could go wrong.

24
01:09.000 --> 01:12.000
And so I had an acronym that's going to help us remember

25
01:12.000 --> 01:14.000
at Jocencies and what's needed to make it happen.

26
01:14.000 --> 01:18.000
And it's STTAMP, like Stamp, with two T's in it.

27
01:18.000 --> 01:22.000
So if you're ever wondering, "Oh, OK, what has to match for those two devices

28
01:22.000 --> 01:25.000
to neighbor up?" think of Stamp.

29
01:25.000 --> 01:26.000
And then we spell them out.

30
01:26.000 --> 01:28.000
The first S is the subnet.

31
01:28.000 --> 01:31.000
They have to agree on what is the network they're both connected to.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

31
01:28.000 --> 01:31.000
They have to agree on what is the network they're both connected to.

32
01:31.000 --> 01:35.440
So if router 1 believes on gig 1, 0, that is directly connected to the 10, 0,

33
01:35.440 --> 01:37.000
12, 0/24 network,

34
01:37.000 --> 01:44.000
which it is, and R2, believes that gig 2, 0 is connected to the 10, 0, 12, 0/25

35
01:44.000 --> 01:44.000
,

36
01:44.000 --> 01:48.440
even though it kind of looks like the same network, it's not the exact same

37
01:48.440 --> 01:49.000
network.

38
01:49.000 --> 01:52.000
If they have to agree on what that common network is between them,

39
01:52.000 --> 01:55.000
if we want to have an OSPF adjacency.

40
01:55.000 --> 01:57.000
The next T is for timers.

41
01:57.000 --> 02:00.000
So in OSPF, they're going to be sending hello messages back and forth.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

42
02:00.000 --> 02:02.000
And there are some timers associated with that.

43
02:02.000 --> 02:06.730
And if the timers for the hello interval and the dead interval, if they don't

44
02:06.730 --> 02:07.000
match,

45
02:07.000 --> 02:09.000
they will not neighbor up.

46
02:09.000 --> 02:11.000
And that's true for IPv4 and IPv6 as well.

47
02:11.000 --> 02:14.000
In fact, all of these apply to both IPv4 and IPv6.

48
02:14.000 --> 02:16.000
And here in the topology, let me give you a lay of the land.

49
02:16.000 --> 02:21.000
I've documented the IP subnets in IPv4, but I'm also using that same number

50
02:21.000 --> 02:25.000
that little x represents the IPv6 subnet address as well.

51
02:25.000 --> 02:33.080
For example, here'd be 2001, DBA6783, 12, and over here'd be 2001, DBA6783, 34,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

51
02:25.000 --> 02:33.080
For example, here'd be 2001, DBA6783, 12, and over here'd be 2001, DBA6783, 34,

52
02:33.080 --> 02:34.000
etc.

53
02:34.000 --> 02:38.180
So for two neighbors to become fully adjacent in OSPF, they need to agree on

54
02:38.180 --> 02:39.000
the subnet.

55
02:39.000 --> 02:41.000
They need to agree on the timers.

56
02:41.000 --> 02:43.000
Another thing they need to agree on is the type.

57
02:43.000 --> 02:45.000
And that is the network type.

58
02:45.000 --> 02:47.000
And let's talk about that here for a moment.

59
02:47.000 --> 02:52.000
On Ethernet, the default network type for OSPF is broadcast.

60
02:52.000 --> 02:55.380
Now, it's funny, I think it's kind of funny, that they don't really use

61
02:55.380 --> 02:56.000
broadcast

62
02:56.000 --> 02:57.000
to find each other.

63
02:57.000 --> 02:58.000
They're using multicast.

64
02:58.000 --> 03:03.000
And those multicast, for IPv4, would be 224.0.0.5.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

64
02:58.000 --> 03:03.000
And those multicast, for IPv4, would be 224.0.0.5.

65
03:03.000 --> 03:08.000
That's an IP address used to reach any OSPF router on that local segment.

66
03:08.000 --> 03:14.000
And 224.0.0.6 is the IP address we'd used to reach and forward information

67
03:14.000 --> 03:19.000
on that local segment to ADR, which also includes a backup designated router.

68
03:19.000 --> 03:24.000
But it's called broadcast with the idea that they can dynamically find each

69
03:24.000 --> 03:24.000
other

70
03:24.000 --> 03:28.000
without specifically pointing in the config to a neighbor's IP address.

71
03:28.000 --> 03:31.000
Other options we're here on the network type would be non-broadcast,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

71
03:28.000 --> 03:31.000
Other options we're here on the network type would be non-broadcast,

72
03:31.000 --> 03:33.000
where it would require additional config.

73
03:33.000 --> 03:37.530
Another option would be point to point, which doesn't use ADR at all on the

74
03:37.530 --> 03:38.000
segment.

75
03:38.000 --> 03:42.000
Another option would be point to multi-point, which also does not use ADR.

76
03:42.000 --> 03:46.000
So in addition to having the timers match, we also need to make sure they agree

77
03:46.000 --> 03:46.000
on,

78
03:46.000 --> 03:50.000
first of all, whether they're expecting to have a DR in the case of broadcast

79
03:50.000 --> 03:53.000
or not have a DR in the case of point to point.

80
03:53.000 --> 03:57.000
So the best practice is to make sure we're using the correct network type.

81
03:57.000 --> 04:00.000
And when we do that, the defaults for the network type is also going to drive
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

83
04:00.000 --> 04:02.000
for the hello and dead intervals.

84
04:02.000 --> 04:04.000
Alright, so we'll make that up a little bit.

85
04:04.000 --> 04:06.000
The A is for area.

86
04:06.000 --> 04:08.000
So they need to agree on a couple of things.

87
04:08.000 --> 04:11.000
They need to agree on the area ID.

88
04:11.000 --> 04:15.000
They also need to agree on the type, the area type.

89
04:15.000 --> 04:20.000
So this is a good chance for us to discuss different area types in OSPF.

90
04:20.000 --> 04:22.000
So let's talk about this topology right here.

91
04:22.000 --> 04:24.000
In fact, let's back it off just a second.

92
04:24.000 --> 04:25.000
Let me clean this up a little bit.

93
04:25.000 --> 04:29.000
And I'm going to go ahead and draw one giant area zero.

94
04:29.000 --> 04:30.000
Another interesting note here.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

95
04:30.000 --> 04:33.000
So instead of area one, it was called the whole thing area zero.

96
04:33.000 --> 04:37.000
So all the interfaces on all the routers are associated with area zero.

97
04:37.000 --> 04:40.000
The challenge here, it will actually, with just eight routers, not a big deal.

98
04:40.000 --> 04:43.000
But let's imagine we had 80 routers.

99
04:43.000 --> 04:48.000
If we had 80 routers, there's little LSAs, link state advertisements

100
04:48.000 --> 04:51.000
that are being generated by these routers.

101
04:51.000 --> 04:54.000
So an LSA type one is called a router LSA.

102
04:54.000 --> 04:58.000
Think of it like a little advertisement that tells all about the router.

103
04:58.000 --> 05:00.000
So our ones jittering at its own LSA type one.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

104
05:00.000 --> 05:03.000
And that's being propagated to all the other routers in the area.

105
05:03.000 --> 05:05.000
Router twos, jittering and LSA type ones.

106
05:05.000 --> 05:08.000
And those LSA type ones are describing which network we're connected to.

107
05:08.000 --> 05:11.000
And the benefit is, is that if every other router in the area

108
05:11.000 --> 05:15.000
sees every other LSA type one, they'll have a perfect picture,

109
05:15.000 --> 05:19.000
like a big puzzle of all the interfaces, how they're connected.

110
05:19.000 --> 05:22.000
And a router can then use that to build this best path information

111
05:22.000 --> 05:25.000
for forwarding packets and add that route then to the routing table.

112
05:25.000 --> 05:28.000
Now if R7 was at a branch office or something,

113
05:28.000 --> 05:32.000
we probably don't want all the LSAs from every single other router
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

113
05:28.000 --> 05:32.000
we probably don't want all the LSAs from every single other router

114
05:32.000 --> 05:36.000
to be forwarded down to R7 because we might be using up some bandwidth.

115
05:36.000 --> 05:41.000
We don't need to use also R7 may not need all that detail about every single

116
05:41.000 --> 05:41.000
other router.

117
05:41.000 --> 05:45.000
So in larger environments where we have dozens or hundreds of routers,

118
05:45.000 --> 05:48.000
instead of using one giant backbone area,

119
05:48.000 --> 05:51.000
we're going to carve it up into multiple different areas.

120
05:51.000 --> 05:54.000
And that would look something like this to pause you right here.

121
05:54.000 --> 05:57.000
Where we have the area zero, the backbone area.

122
05:57.000 --> 06:01.000
So the backbone area on our one, we've associated gig zero zero,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

122
05:57.000 --> 06:01.000
So the backbone area on our one, we've associated gig zero zero,

123
06:01.000 --> 06:06.000
one zero two zero, and our two gig two zero, one zero, and our four gig two

124
06:06.000 --> 06:06.000
zero,

125
06:06.000 --> 06:11.000
FA four slash one, and our three FA four slash zero, and gig one zero.

126
06:11.000 --> 06:14.000
And we've associated all those with area zero.

127
06:14.000 --> 06:16.000
Now one of the rules, if you want to have multiple areas,

128
06:16.000 --> 06:18.000
for example, area one here and area two here,

129
06:18.000 --> 06:23.300
we need to have at least one router that's connecting the non backbone areas to

130
06:23.300 --> 06:24.000
the backbone.

131
06:24.000 --> 06:26.000
So let's start here with area one.

132
06:26.000 --> 06:31.000
Router three has two interfaces associated with area zero, and if we associate
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

132
06:26.000 --> 06:31.000
Router three has two interfaces associated with area zero, and if we associate

133
06:31.000 --> 06:32.000
this interface right here,

134
06:32.000 --> 06:35.000
FA four slash one with area one, boom!

135
06:35.000 --> 06:39.000
The begin of our three act as a gateway between all the networks here,

136
06:39.000 --> 06:43.000
and forwarding traffic either to the backbone area or on two area two.

137
06:43.000 --> 06:48.000
So here on our five, this interface, we assign to area one, area one there two,

138
06:48.000 --> 06:49.000
area one, and area one.

139
06:49.000 --> 06:52.000
And again, as the interfaces are associated with the area,

140
06:52.000 --> 06:55.000
not the actual router itself, because our one is a backbone router,

141
06:55.000 --> 06:58.000
because all of its interfaces are in area zero.

142
06:58.000 --> 07:01.000
Our seven and our five have all their interfaces in area one,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

142
06:58.000 --> 07:01.000
Our seven and our five have all their interfaces in area one,

143
07:01.000 --> 07:05.350
but this guy, R three, acting as the border router between the backbone area

144
07:05.350 --> 07:06.000
and area one,

145
07:06.000 --> 07:10.000
it has a special name, as a result of having interfaces in the backbone,

146
07:10.000 --> 07:14.000
and also in area one, and it's called an area border router.

147
07:14.000 --> 07:18.290
And with an area border router, what it does is going to generate LSA type

148
07:18.290 --> 07:19.000
threes.

149
07:19.000 --> 07:24.000
And the name for the LSA type three coming from the ABR is a summary LSA.

150
07:24.000 --> 07:27.000
And the benefit of a summary LSA is that it has information, for example,

151
07:27.000 --> 07:32.000
about the 10 zero 13 network and 10 zero 12 and 10 zero 24 and 10 zero 34,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

151
07:27.000 --> 07:32.000
about the 10 zero 13 network and 10 zero 12 and 10 zero 24 and 10 zero 34,

152
07:32.000 --> 07:36.000
and also information about the network 10246, etc.

153
07:36.000 --> 07:40.000
And it can forward all those individual routes down into area one.

154
07:40.000 --> 07:43.000
So our seven and our five can know about those routes,

155
07:43.000 --> 07:46.000
but it's kind of a summary in the sense that it doesn't have to know,

156
07:46.000 --> 07:49.000
these routers don't have to know all the details about the LSA type ones,

157
07:49.000 --> 07:51.000
and all the other details about those routers.

158
07:51.000 --> 07:54.750
All they know is that those routes are reachable through the area border router

159
07:54.750 --> 07:55.000
.

160
07:55.000 --> 07:59.000
So in this case, our seven, if we had, for example, 80 routers

161
07:59.000 --> 08:01.000
and most of those routers were in different areas,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

161
07:59.000 --> 08:01.000
and most of those routers were in different areas,

162
08:01.000 --> 08:06.000
our seven would not have to see or have propagated the LSA type ones

163
08:06.000 --> 08:10.000
from any of these backbone routers like R1, R2, R4,

164
08:10.000 --> 08:13.000
or any of the other dozens or hundreds of routers,

165
08:13.000 --> 08:15.000
because it's in a different area.

166
08:15.000 --> 08:20.000
That's one of the benefits of chopping up OSPF into logical areas or sub areas.

167
08:20.000 --> 08:24.000
Also, we're talking about LSAs. Let's also add an LSA type two.

168
08:24.000 --> 08:28.000
And an LSA type two is referred to as a network LSA.

169
08:28.000 --> 08:31.000
And that's generated by a designated router on a segment.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

169
08:28.000 --> 08:31.000
And that's generated by a designated router on a segment.

170
08:31.000 --> 08:34.000
So for example, let's go over here to R2 and R4.

171
08:34.000 --> 08:37.000
One of these two routers is going to be the designated router

172
08:37.000 --> 08:40.000
for the 10024 network for the segment.

173
08:40.000 --> 08:42.000
So I'm going to put in parentheses, DR,

174
08:42.000 --> 08:45.000
generates that network LSA that describes that network.

175
08:45.000 --> 08:47.000
So this is the type right here.

176
08:47.000 --> 08:50.000
So the LSA type one is generated by each router.

177
08:50.000 --> 08:54.240
By setting up multiple areas, we're controlling the reach and span of those LSA

178
08:54.240 --> 08:55.000
type ones.

179
08:55.000 --> 08:59.300
So for example, the LSA type ones from R6 are never going to be advertised here

180
08:59.300 --> 09:00.000
in area one,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

181
09:00.000 --> 09:02.000
because they're completely different area.

182
09:02.000 --> 09:06.000
And that frees up CPU and resources on R5 and other routers in this area,

183
09:06.000 --> 09:09.000
because they don't have to process it or look at that information.

184
09:09.000 --> 09:12.000
Same thing about the DRs coming from a different area.

185
09:12.000 --> 09:14.000
So for an LSA type two, for example,

186
09:14.000 --> 09:17.410
the great is the designated router is generating an LSA type two describing

187
09:17.410 --> 09:18.000
that network.

188
09:18.000 --> 09:22.000
That LSA type two is never going to be propagated over into area one,

189
09:22.000 --> 09:25.000
because area one doesn't need that information.

190
09:25.000 --> 09:28.000
Because it's getting summary route information from the ABR

191
09:28.000 --> 09:31.000
that describes how to get to all those given networks
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

191
09:28.000 --> 09:31.000
that describes how to get to all those given networks

192
09:31.000 --> 09:34.000
without the additional burden of processing the individual LSAs.

193
09:34.000 --> 09:37.000
Think of it like less information is more.

194
09:37.000 --> 09:41.000
So the summary route are advertised by area border routers.

195
09:41.000 --> 09:44.000
And those are advertised for example, R3 would be advertising summaries.

196
09:44.000 --> 09:50.000
Think of it like the information in the routes for a 13, 12, 24, 34, etc.

197
09:50.000 --> 09:52.000
And it would be advertising those routes,

198
09:52.000 --> 09:55.720
but it wouldn't be forwarding down the LSA type ones and type twos into area

199
09:55.720 --> 09:56.000
one

200
09:56.000 --> 09:58.000
from any of those other areas.

201
09:58.000 --> 10:00.000
Because these routers here don't need that information,
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

202
10:00.000 --> 10:02.000
because they have the summary routes.

203
10:02.000 --> 10:04.000
So that leads me to the area type.

204
10:04.000 --> 10:06.000
And as we've described it so far,

205
10:06.000 --> 10:09.000
area zero here would be a normal area, the backbone area,

206
10:09.000 --> 10:12.000
area one would be a normal area, area two would be a normal area.

207
10:12.000 --> 10:15.000
So what exactly is an unnormal area?

208
10:15.000 --> 10:18.000
So as far as areas go, we'll jot down normal.

209
10:18.000 --> 10:21.000
What normal means is that we could take an external route.

210
10:21.000 --> 10:25.000
For example, EIGRP or BGP or Static Route.

211
10:25.000 --> 10:29.000
And we could redistribute it into OSPF in that area.

212
10:29.000 --> 10:32.000
That's an example of what we could do in a normal area.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

212
10:29.000 --> 10:32.000
That's an example of what we could do in a normal area.

213
10:32.000 --> 10:35.000
And so we might have that feature and that ability and need

214
10:35.000 --> 10:37.000
to do that in the backbone area.

215
10:37.000 --> 10:40.000
However, let's imagine that area one and area two are being used

216
10:40.000 --> 10:42.000
in two separate geographical locations.

217
10:42.000 --> 10:45.000
And we don't need to do redistribution in those areas.

218
10:45.000 --> 10:46.000
We want to simplify it.

219
10:46.000 --> 10:49.000
What we could do is we can make those areas a stub area.

220
10:49.000 --> 10:52.000
So a stub area says, okay, we're not allowing the redistribution

221
10:52.000 --> 10:55.000
of other routes into OSPF.

222
10:55.000 --> 10:58.000
For example, from BGP or from EIGRP or from Static Route.

223
10:58.000 --> 10:59.000
Which is great.

224
10:59.000 --> 11:02.000
And there's also an option of something called a totally stubby area.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

224
10:59.000 --> 11:02.000
And there's also an option of something called a totally stubby area.

225
11:02.000 --> 11:05.000
So let's imagine we made these areas a totally stubby area.

226
11:05.000 --> 11:06.000
Both of them.

227
11:06.000 --> 11:08.000
What that means is these two routers are three and are four,

228
11:08.000 --> 11:11.000
which are the area border routers for area one and two respectively.

229
11:11.000 --> 11:15.000
With a totally stubby area, they are not going to forward

230
11:15.000 --> 11:17.000
in even the summary routes.

231
11:17.000 --> 11:22.000
So area one here and the routers here wouldn't learn about 10-012 and 10-034.

232
11:22.000 --> 11:23.000
Anything else?

233
11:23.000 --> 11:26.000
Because the ABR and a totally stubby area is not sending those routes.

234
11:26.000 --> 11:29.000
Now the problem with a totally stubby area is that if we are not receiving

235
11:29.000 --> 11:34.000
a far-fired has no idea about the 10-012 and 10-024 network and other networks.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

235
11:29.000 --> 11:34.000
a far-fired has no idea about the 10-012 and 10-024 network and other networks.

236
11:34.000 --> 11:37.690
How in the world of these routers going to know how to forward if they need to

237
11:37.690 --> 11:39.000
forward a packet to any of those networks.

238
11:39.000 --> 11:44.580
And the answer is in a totally stubby area, we can also advertise a default

239
11:44.580 --> 11:46.000
route into OSPF.

240
11:46.000 --> 11:48.000
So that way, our five and our seven.

241
11:48.000 --> 11:50.000
Even though they're not getting these summary routes,

242
11:50.000 --> 11:52.000
we're getting all the individual routes from the other OSPF areas.

243
11:52.000 --> 11:56.550
They are getting a default route and then they can forward packets using that

244
11:56.550 --> 11:58.000
default route in the direction of the ABR.

245
11:58.000 --> 12:01.690
And again, that's going to reduce the number of LSAs here in area one and area
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

245
11:58.000 --> 12:01.690
And again, that's going to reduce the number of LSAs here in area one and area

246
12:01.690 --> 12:03.000
two through both totally stubby

247
12:03.000 --> 12:06.000
because they're now not receiving LSA type ones from other areas.

248
12:06.000 --> 12:09.000
They're not receiving LSA type two's from other areas.

249
12:09.000 --> 12:14.130
And they're not receiving LSA type three summary LSAs with the exception of the

250
12:14.130 --> 12:17.000
default route that's being injected into the area by the ABR.

251
12:17.000 --> 12:20.970
So why the big discussion Keith regarding types? Well the answer is that type

252
12:20.970 --> 12:22.000
has to match.

253
12:22.000 --> 12:29.000
If this interface here thinks it's a normal area and this interface here thinks

254
12:29.000 --> 12:30.000
it's a totally stubby area
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

255
12:30.000 --> 12:34.170
that will prevent them from neighboring up. So that's why the area D, for

256
12:34.170 --> 12:37.000
example area one, has to match and the type.

257
12:37.000 --> 12:40.000
In this example, just a stub area or a normal area.

258
12:40.000 --> 12:43.500
Whatever we set up, it has to match between those two interfaces on that common

259
12:43.500 --> 12:44.000
segment.

260
12:44.000 --> 12:48.120
And there's also an option, first thing called a not-so-stubby area, which is

261
12:48.120 --> 12:50.000
where you want to have this be like a stub area

262
12:50.000 --> 12:53.970
but you want to make an exception to be able to inject or redistribute routes

263
12:53.970 --> 12:55.000
into that area.

264
12:55.000 --> 12:58.310
So for further discussion we're getting configuring and working with and

265
12:58.310 --> 13:01.000
playing with stub and totally stubby and not-so-stubby areas.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

265
12:58.310 --> 13:01.000
playing with stub and totally stubby and not-so-stubby areas.

266
13:01.000 --> 13:05.270
We're going to say those discussions as part of a skill in another course, very

267
13:05.270 --> 13:06.000
likely an RC.

268
13:06.000 --> 13:09.570
But I want you to be aware of some of the options here because the type has to

269
13:09.570 --> 13:14.000
match if we want to neighbor up with another device on that same segment.

270
13:14.000 --> 13:18.000
So back to our pneumonic of stamp. The M is for M-TU.

271
13:18.000 --> 13:21.900
Now if you have two routers the same model or similar models, it's very likely

272
13:21.900 --> 13:25.000
the M-TU for Ethernet is going to be 1500 bytes.

273
13:25.000 --> 13:29.440
No problem. However, let's say we're peering up with a multi-layer switch and

274
13:29.440 --> 13:31.000
Cisco catalyst, whatever.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

274
13:29.440 --> 13:31.000
Cisco catalyst, whatever.

275
13:31.000 --> 13:36.680
And the M-TU on the router is 1500 and the multi-layer switch, the M-TU is 1504

276
13:36.680 --> 13:37.000
.

277
13:37.000 --> 13:42.000
That will cause a problem if the M-TU isn't the same size.

278
13:42.000 --> 13:45.000
And that's kind of frustrating because you can ping and connect back and forth.

279
13:45.000 --> 13:48.000
However, they won't neighbor up if the M-TU doesn't agree.

280
13:48.000 --> 13:53.350
So we do their manipulate the M-TU here and make it match 1500 or we can

281
13:53.350 --> 13:58.000
configure OSPF to ignore the M-TU and simply neighbor up anyway.

282
13:58.000 --> 14:03.000
And then the P is for password. Now I use password so the pneumonic would work.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

282
13:58.000 --> 14:03.000
And then the P is for password. Now I use password so the pneumonic would work.

283
14:03.000 --> 14:05.000
However, really what that is is authentication.

284
14:05.000 --> 14:09.210
So the authentication method between two routers on the same segment that has

285
14:09.210 --> 14:10.000
to match.

286
14:10.000 --> 14:13.870
So for R1 and R3 to neighbor up, they need to be specifying using the same type

287
14:13.870 --> 14:15.000
of authentication.

288
14:15.000 --> 14:19.430
So if one is set for plain text and they're set to M-D5, that's not going to

289
14:19.430 --> 14:20.000
fly.

290
14:20.000 --> 14:22.770
And further, if they agree on the type of authentication, we also need to make

291
14:22.770 --> 14:24.000
sure if we're using a password,

292
14:24.000 --> 14:27.940
that password matches on both sides or in the case of M-D5, the hash of that

293
14:27.940 --> 14:30.000
password matches on both sides.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

294
14:30.000 --> 14:34.000
Alright, so let's clean that up and let's take a look at the metrics involved

295
14:34.000 --> 14:36.000
with OSPF and how it decides on the best path.

296
14:36.000 --> 14:41.000
For the path selection for OSPF, it uses a metric and that metric is cost.

297
14:41.000 --> 14:46.380
And just like going to shop at a store, if we had two comparable products, they

298
14:46.380 --> 14:48.000
were, you know, versatile identical,

299
14:48.000 --> 14:51.680
except the cost was different. We choose the lowest cost. And that's what OSPF

300
14:51.680 --> 14:52.000
does.

301
14:52.000 --> 14:56.450
So each of our interfaces has an associated cost. Now by default, it's

302
14:56.450 --> 14:59.000
unfortunate, but it's kind of a legacy thing.

303
14:59.000 --> 15:02.000
There is something called an auto cost reference bandwidth.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

303
14:59.000 --> 15:02.000
There is something called an auto cost reference bandwidth.

304
15:02.000 --> 15:05.800
We've talked about that in our CCNAG courses and it's also relevant here

305
15:05.800 --> 15:10.000
because the default is 100 megabits per second.

306
15:10.000 --> 15:14.340
And a few decades ago, it wasn't a problem using that value when we had much

307
15:14.340 --> 15:16.000
slower speed networks.

308
15:16.000 --> 15:19.000
However, the auto cost reference bandwidth works like this.

309
15:19.000 --> 15:22.000
We take the reference bandwidth and we take our actual interface bandwidth.

310
15:22.000 --> 15:25.180
Let's take a fast ethernet, which is 100 megabits per second in the case of

311
15:25.180 --> 15:26.000
fast ethernet.

312
15:26.000 --> 15:30.000
So fast ethernet, if we divide 100 into 100 equals 1.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

313
15:30.000 --> 15:34.440
So that would have a cost of 1. Well, let's go ahead and do that same math with

314
15:34.440 --> 15:35.000
gigabit.

315
15:35.000 --> 15:39.930
So the reference bandwidth again, if it's 100 and the interface bandwidth is

316
15:39.930 --> 15:43.000
1000, meaning 1000 megabits per second,

317
15:43.000 --> 15:48.000
1000 going into 100 is 0.1. However, we're going to round it up.

318
15:48.000 --> 15:51.290
So for gigabit ethernet by default, if we're using the default reference

319
15:51.290 --> 15:54.000
bandwidth, the cost of a gigabit link is 1.

320
15:54.000 --> 15:57.830
So it kind of makes OSPF blind to the difference between fast ethernet and

321
15:57.830 --> 15:59.000
anything faster.

322
15:59.000 --> 16:03.000
So ideally, what we want to do is increase the reference bandwidth.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

322
15:59.000 --> 16:03.000
So ideally, what we want to do is increase the reference bandwidth.

323
16:03.000 --> 16:07.560
So if in this apology, we could set it to 1000 and that would make our gigabit

324
16:07.560 --> 16:09.000
links look like a cost of 1.

325
16:09.000 --> 16:12.000
And our fast ethernet looked like a cost of 10.

326
16:12.000 --> 16:15.300
However, in environments where you have 10 gig, 40 gig, and higher, you may

327
16:15.300 --> 16:17.000
want to set it to even higher than that.

328
16:17.000 --> 16:20.000
So in the lab environment, that's what we're going to be doing consistently.

329
16:20.000 --> 16:23.000
We're going to set the auto cost reference bandwidth to 1000.

330
16:23.000 --> 16:26.000
And that way, gigabit links will look like a cost of 1.

331
16:26.000 --> 16:29.000
Fasted it looked like a cost of 10. I document that real quick.

332
16:29.000 --> 16:33.510
And these serial links, I have set at 64. The bandwidth set to 64 kilobits per
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

332
16:29.000 --> 16:33.510
And these serial links, I have set at 64. The bandwidth set to 64 kilobits per

333
16:33.510 --> 16:34.000
second.

334
16:34.000 --> 16:37.000
And so their value for these serial links, actually, we're not using this one.

335
16:37.000 --> 16:41.000
And we're not using this one. But I do have OSPF enabled on these two.

336
16:41.000 --> 16:46.000
So the cost of that serial link, I'll put it right here, is 15,625.

337
16:46.000 --> 16:49.890
And we can see that once OSPF is configured with the show IP, OSPF interface

338
16:49.890 --> 16:51.000
brief for IPv4.

339
16:51.000 --> 16:55.380
And so if we are going to calculate, for example, from the perspective of our

340
16:55.380 --> 16:56.000
eight,

341
16:56.000 --> 17:01.000
it's cost to get to the 10.01 network, assuming the network's converged.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

341
16:56.000 --> 17:01.000
it's cost to get to the 10.01 network, assuming the network's converged.

342
17:01.000 --> 17:04.000
And all our neighborships are set up for this target network.

343
17:04.000 --> 17:08.000
The OSPF cost should be 11 here. So you 1 plus.

344
17:08.000 --> 17:12.000
And then 15,625 plus.

345
17:12.000 --> 17:15.000
And the best path would be, then the cost be 1 here.

346
17:15.000 --> 17:18.000
And then going through our two, the cost is 1 there.

347
17:18.000 --> 17:21.000
And then we count the final network, which is 1 here.

348
17:21.000 --> 17:24.000
So it's all the egress interfaces that we're looking at.

349
17:24.000 --> 17:31.330
So going out here for 1, out here for 15,625, here for 1, here for 1, and here
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

349
17:24.000 --> 17:31.330
So going out here for 1, out here for 15,625, here for 1, here for 1, and here

350
17:31.330 --> 17:32.000
for 1.

351
17:32.000 --> 17:33.000
And that's our final destination.

352
17:33.000 --> 17:37.170
So if all those values are correct, that would be a cost from our eights

353
17:37.170 --> 17:38.000
perspective

354
17:38.000 --> 17:42.000
to get to the 10.01 network using the best or optimized path.

355
17:42.000 --> 17:45.000
That'd be 15,629.

356
17:45.000 --> 17:49.030
Now one of the cool things about this set of videos and the skill is that I

357
17:49.030 --> 17:50.000
also have a hands-on lab for you.

358
17:50.000 --> 17:54.000
So in addition to predicting like what we think the cost is going to be,

359
17:54.000 --> 17:58.000
we can also once it's labbed up, validate that that indeed is the case.

360
17:58.000 --> 18:01.550
And then the event that we have two or more routes with equal cost paths of the
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

360
17:58.000 --> 18:01.550
And then the event that we have two or more routes with equal cost paths of the

361
18:01.550 --> 18:02.000
OSPF,

362
18:02.000 --> 18:05.000
it can use equal cost load balancing.

363
18:05.000 --> 18:08.450
Now one more piece of information I like to share in this overview is that

364
18:08.450 --> 18:12.000
there is more than just one way to configure OSPF.

365
18:12.000 --> 18:17.000
Meaning we can use the traditional method for setting up OSPF for IPv4.

366
18:17.000 --> 18:20.000
There's also a traditional method for IPv6.

367
18:20.000 --> 18:27.490
And we also can use a OSPF V3 option for configuring both IPv4 and IPv6, or

368
18:27.490 --> 18:32.000
either IPv4 or IPv6 using the concept of address families.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

368
18:27.490 --> 18:32.000
either IPv4 or IPv6 using the concept of address families.

369
18:32.000 --> 18:35.490
Not too different from actually how we configured it with eAGRP in a previous

370
18:35.490 --> 18:36.000
skill.

371
18:36.000 --> 18:39.710
So we're going to have a separate set of videos on configuring OSPF version 3

372
18:39.710 --> 18:41.000
and working with it.

373
18:41.000 --> 18:45.120
So for this set of videos, I'm going to actually walk you through examples of

374
18:45.120 --> 18:51.360
configuring IPv4 traditionally and IPv6 with OSPF using this topology with the

375
18:51.360 --> 18:53.000
interface is shown in green here in area 0.

376
18:53.000 --> 18:56.000
And the interface is shown in blue here in area 1.

377
18:56.000 --> 18:59.430
And the interface is over here in whatever this color is, kind of a pinkish

378
18:59.430 --> 19:00.000
thing in area 2.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

379
19:00.000 --> 19:04.440
And as we go through this together as part of this set of videos, you'll have

380
19:04.440 --> 19:07.190
the opportunity of seeing how it's configured and also doing the configuration

381
19:07.190 --> 19:08.000
of the hands on lab.

382
19:08.000 --> 19:13.090
But most importantly, regarding understanding OSPF is we can actually see and

383
19:13.090 --> 19:15.000
take a look at the results.

384
19:15.000 --> 19:17.000
We can take a look at the LSAs.

385
19:17.000 --> 19:20.600
We can practice with the show commands to verify that what we think is

386
19:20.600 --> 19:23.000
happening is indeed happening with OSPF.

387
19:23.000 --> 19:26.700
So with that in mind, in the next video, I like to demonstrate for you the

388
19:26.700 --> 19:31.850
implementation of OSPF with IPv4 using the traditional option and using three
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

388
19:26.700 --> 19:31.850
implementation of OSPF with IPv4 using the traditional option and using three

389
19:31.850 --> 19:32.000
areas.

390
19:32.000 --> 19:37.150
The backbone area, area 0, and then area 1 and area 2 connected to the backbone

391
19:37.150 --> 19:42.000
through area border routers.