1
00:00:02,241 --> 00:00:08,657
[music]

2
00:00:08,657 --> 00:00:15,985
So let's jump in with just
standard GRE tunnels over IPv6.

3
00:00:15,985 --> 00:00:19,422
For this, we're going to be
using a different topology.

4
00:00:19,422 --> 00:00:25,434
This is also saved out there
under IPv6 VPN topology.

5
00:00:25,434 --> 00:00:28,603
The basic idea here
is, we're going

6
00:00:28,603 --> 00:00:32,200
to use this for VPN MPLS,

7
00:00:32,200 --> 00:00:37,943
really for all of our tunneling
or VPN technologies here.

8
00:00:37,943 --> 00:00:40,307
We're going to start with GRE,
and we're not going to use all

9
00:00:40,307 --> 00:00:42,323
of these devices for
all of the sections.

10
00:00:42,323 --> 00:00:44,199
There's really no need.

11
00:00:44,199 --> 00:00:46,480
But this is the
topology that's built

12
00:00:46,480 --> 00:00:48,541
that will support everything

13
00:00:48,541 --> 00:00:51,741
we need it to support
moving forward.

14
00:00:51,741 --> 00:00:54,283
The important thing
to understand here

15
00:00:54,283 --> 00:01:00,944
is that the switches are just
acting like any devices.

16
00:01:00,944 --> 00:01:04,546
We have the IPv4
address listed here,

17
00:01:04,546 --> 00:01:07,583
 but it corresponds with what

18
00:01:07,583 --> 00:01:12,338
the IPv6 address is going to
be on that subnet as well,

19
00:01:12,338 --> 00:01:15,885
and you'll see those as we go.

20
00:01:15,885 --> 00:01:24,076
VLANs 71, 91, 84, 65, 105, these
are all IPv6 enabled already.

21
00:01:24,076 --> 00:01:28,538
Now, our PE routers,
which again,

22
00:01:28,538 --> 00:01:29,830
it may or may not be provided,

23
00:01:29,830 --> 00:01:32,911
could of course also just
be your edge routers.

24
00:01:32,911 --> 00:01:34,410
That's irrelevant.

25
00:01:34,410 --> 00:01:36,775
We're just looking at
the technologies here.

26
00:01:36,775 --> 00:01:41,689
But those are also already
configured with IPv6 addresses.

27
00:01:41,689 --> 00:01:45,382
But as we move into technologies
such as MPLS and such,

28
00:01:45,382 --> 00:01:47,773
we will be redoing.

29
00:01:47,773 --> 00:01:52,047
So we're going to
start with just GRE,

30
00:01:52,047 --> 00:01:53,664
and we'll just build one tunnel

31
00:01:53,664 --> 00:01:55,793
between R1 and R4.

32
00:01:55,793 --> 00:02:01,613
So essentially, we're just going
to get our IPv6 from switch 1 to

33
00:02:01,613 --> 00:02:07,555
switch 2 across the top there,
and tunnel between R1 and R4.

34
00:02:07,555 --> 00:02:09,549
The most important thing to note

35
00:02:09,549 --> 00:02:15,476
about this diagram
is that the core

36
00:02:15,476 --> 00:02:17,881
in the middle there,
everything inside the cloud

37
00:02:17,881 --> 00:02:21,230
if you will, is IPv4 only.

38
00:02:21,230 --> 00:02:23,880
It's going to remain
that way the whole time.

39
00:02:23,880 --> 00:02:28,304
There's no IPv6 in the cloud.

40
00:02:28,304 --> 00:02:35,203
As we move into this, let's hop
over here to our command line.

41
00:02:35,203 --> 00:02:36,180
Let's take a look.

42
00:02:36,180 --> 00:02:46,906
On router 1, for example, if I
were to say do show ipv6 route,

43
00:02:46,906 --> 00:02:52,193
all he has is literally
a route to Null0.

44
00:02:52,193 --> 00:02:55,213
That's it. He has
no IPv6 routes.

45
00:02:55,213 --> 00:02:59,327
If I said do show
ipv6 interface brief,

46
00:02:59,327 --> 00:03:02,884
he going to have
addresses, like I said,

47
00:03:02,884 --> 00:03:07,352
on his 71 and 91
interfaces going out

48
00:03:07,352 --> 00:03:10,090
to the switches,
but not anything

49
00:03:10,090 --> 00:03:12,667
on the core. His core interface

50
00:03:12,667 --> 00:03:17,176
is Fa0/0.12 and that is
currently unassigned.

51
00:03:17,176 --> 00:03:20,938
Also, if I said
do show ip route,

52
00:03:20,938 --> 00:03:26,745
I want you to note that the only
IPv4 routes he has is to either

53
00:03:26,745 --> 00:03:32,141
connected interfaces or the
other device's loopbacks.

54
00:03:32,141 --> 00:03:35,379
The loopback numeration
is on the diagram there.

55
00:03:35,379 --> 00:03:38,199
It's a little bit different
than what we did before.

56
00:03:38,199 --> 00:03:42,711
It's just the device
numbers all the way out.

57
00:03:42,711 --> 00:03:46,630
Just to jump in, the first thing
we're going to do is we need to

58
00:03:46,630 --> 00:03:51,868
fire up-- need to is a
bad word, I suppose.

59
00:03:51,868 --> 00:03:53,657
We could alway do static routes,

60
00:03:53,657 --> 00:03:57,252
but we're going to fire up

61
00:03:57,252 --> 00:04:00,786
a routing protocol
over all of this.

62
00:04:00,786 --> 00:04:08,291
We're going to say-- in this
case, we'll say, Rack1R1.

63
00:04:08,291 --> 00:04:12,952
We're going to say,
interface tunnel 0,

64
00:04:12,952 --> 00:04:16,863
and we will say,
tunnel-- well, I

65
00:04:16,863 --> 00:04:17,866
just said about the protocol and

66
00:04:17,866 --> 00:04:19,977
I'm jumping into the tunnel.

67
00:04:19,977 --> 00:04:22,032
Let's do our protocols first,

68
00:04:22,032 --> 00:04:23,428
although we could turn
it on here as well

69
00:04:23,428 --> 00:04:24,501
since we're already here.

70
00:04:24,501 --> 00:04:29,116
We'll just do EIGRP, just
to pick one to start with.

71
00:04:29,116 --> 00:04:37,419
So ipv6 eigrp 100 - easy
enough - interface.

72
00:04:37,419 --> 00:04:38,761
And like I said,
this is just going

73
00:04:38,761 --> 00:04:40,265
to be the one up at the top.

74
00:04:40,265 --> 00:04:48,578
So this is going to be
interface FA0/0.71,

75
00:04:48,578 --> 00:04:51,532
also running EIGRP 100.

76
00:04:51,532 --> 00:04:56,537
And then ipv6 router eigrp 100.

77
00:04:56,537 --> 00:04:58,713
That starts the global process.

78
00:04:58,713 --> 00:05:06,434
Do show ipv6 eigrp
interface, and it's 71.

79
00:05:06,434 --> 00:05:07,074
Perfect.

80
00:05:10,007 --> 00:05:16,047
We can do the exact same thing
over on router 4 as well.

81
00:05:16,047 --> 00:05:18,031
So if you'll take a look
at the diagram here,

82
00:05:18,031 --> 00:05:22,321
remember we're going from
router 1 to router 4,

83
00:05:22,321 --> 00:05:26,720
and then of course it will
involve VLAN 71 and VLAN 84.

84
00:05:26,720 --> 00:05:29,349
That's all we're going to
work on for right now.

85
00:05:29,349 --> 00:05:30,889
So I just want to
make sure everyone

86
00:05:30,889 --> 00:05:32,410
understands where we're at.

87
00:05:32,410 --> 00:05:42,552
On router 4, interface
tunnel 0, ipv6 eigrp 100,

88
00:05:42,552 --> 00:05:48,063
interface fa0/0.84.

89
00:05:53,514 --> 00:05:59,508
Also in EIGRP 100
and then router

90
00:05:59,508 --> 00:06:06,362
ipV6 router eigrp 100,
and then our switches.

91
00:06:06,362 --> 00:06:15,842
So switch 1, interface
loopback 0,

92
00:06:15,842 --> 00:06:19,638
ipv6 eigrp 100.

93
00:06:19,638 --> 00:06:22,247
Loopback 1, by the
way, is the one

94
00:06:22,247 --> 00:06:24,810
that has the, we'll call it the

95
00:06:24,810 --> 00:06:26,313
client prefix.

96
00:06:26,313 --> 00:06:28,041
Let me just show you
here real quick.

97
00:06:28,041 --> 00:06:33,299
Do show ipv6 interface brief.

98
00:06:33,299 --> 00:06:40,600
If you look at his
loopback 1, this is a /64.

99
00:06:40,600 --> 00:06:44,719
This is like the client subnet

100
00:06:44,719 --> 00:06:46,635
that this switch is presenting,

101
00:06:46,635 --> 00:06:49,787
and then this is his normal
management loopback,

102
00:06:49,787 --> 00:06:52,521
if you will. I'm going to
bring them both in EIGRP.

103
00:06:52,521 --> 00:06:53,996
It doesn't really  matter.

104
00:06:53,996 --> 00:06:59,240
I really just want the loopback
1, but we'll bring them all in.

105
00:06:59,240 --> 00:07:04,568
Then of course, router
ipv6 eigrp 100.

106
00:07:04,568 --> 00:07:05,843
That's backwards.

107
00:07:17,654 --> 00:07:19,258
 Then switch 2.

108
00:07:22,022 --> 00:07:22,965
Here's the good thing we can do.

109
00:07:22,965 --> 00:07:26,169
We can just say-- well,
I had typos in there.

110
00:07:26,169 --> 00:07:27,175
It's not even going
to help us really.

111
00:07:27,175 --> 00:07:29,650
Switch 2 doesn't really matter.

112
00:07:29,650 --> 00:07:31,478
Interface loopback 0,

113
00:07:31,478 --> 00:07:34,598
ipv6 eigrp 100,

114
00:07:34,598 --> 00:07:36,688
interface loopback 1.

115
00:07:36,688 --> 00:07:43,487
eigrp 100. Interface
vlan, and this is 84,

116
00:07:43,487 --> 00:07:50,592
eigrp 100, ipv6
router eigrp 100.

117
00:07:50,592 --> 00:07:53,302
Neighbors should come up.

118
00:07:53,302 --> 00:07:54,726
So 2 and 4 are up,

119
00:07:54,726 --> 00:08:03,777
although I did not see switch
1 and R1 never came up.

120
00:08:03,777 --> 00:08:08,041
Never did the VLAN interface.

121
00:08:08,041 --> 00:08:10,065
Interface vlan 71.

122
00:08:14,339 --> 00:08:17,398
There you go. Neighbors
should come up.

123
00:08:17,398 --> 00:08:19,608
So if we've done
our job correctly,

124
00:08:19,608 --> 00:08:21,381
we should be able to
go to router 1 now

125
00:08:21,381 --> 00:08:25,553
and say do show
ipv6 route eigrp,

126
00:08:25,553 --> 00:08:28,974
and he should have router 1's
information, which he does.

127
00:08:28,974 --> 00:08:32,262
We should be able to
go over to router 4,

128
00:08:32,262 --> 00:08:38,875
do show ipv6 route eigrp, and he

129
00:08:38,875 --> 00:08:45,559
should have switch
2's information,

130
00:08:45,559 --> 00:08:46,834
which he does.

131
00:08:46,834 --> 00:08:47,803
So that's good.

132
00:08:47,803 --> 00:08:51,526
Now we go back to our tunnel.

133
00:08:51,526 --> 00:08:59,229
Interface tunnel 0, and
we say ipv6 address.

134
00:08:59,229 --> 00:09:09,142
I don't know 2001 DB8:100:200::,
and he's the 4 end of that, /64.

135
00:09:09,142 --> 00:09:12,272
So we put an IPv6
address on the tunnel.

136
00:09:12,272 --> 00:09:13,966
We can do that on both sides.

137
00:09:28,690 --> 00:09:37,683
Then tunnel source is
going to be loopback 0.

138
00:09:37,683 --> 00:09:41,466
Tunnel destination is 4.4.4.4.

139
00:09:41,466 --> 00:09:45,242
That's router 4's address.

140
00:09:45,242 --> 00:09:46,549
It's going to say that it's up.

141
00:09:46,549 --> 00:09:47,790
We know that's nonsense.

142
00:09:47,790 --> 00:09:49,251
It's a false up.

143
00:09:49,251 --> 00:09:53,519
If you want, you can
say like keep 2,

144
00:09:53,519 --> 00:09:55,186
and send keepalives
every 2 seconds,

145
00:09:55,186 --> 00:09:56,552
then within six
seconds, it should

146
00:09:56,552 --> 00:09:58,832
tell me the tunnel's
down because

147
00:09:58,832 --> 00:10:00,510
there are no keepalives
on the other side.

148
00:10:00,510 --> 00:10:02,351
These are GRE keepalives.

149
00:10:02,351 --> 00:10:03,221
There we go.

150
00:10:03,221 --> 00:10:04,199
Tunnel goes down.

151
00:10:04,199 --> 00:10:05,986
That's when it
realizes, Oh wait,

152
00:10:05,986 --> 00:10:08,647
the other end's not really up.

153
00:10:08,647 --> 00:10:12,779
The only thing this up meant,
just like any other GRE,

154
00:10:12,779 --> 00:10:18,548
is that you have a route to get
to the tunnel destination.

155
00:10:18,548 --> 00:10:20,913
It's all it really means.

156
00:10:20,913 --> 00:10:26,766
So router 4, tunnel
source loopback0,

157
00:10:26,766 --> 00:10:29,060
tunnel destination 1.1.1.1.

158
00:10:33,560 --> 00:10:36,988
That's like IPV
geometry or something.

159
00:10:36,988 --> 00:10:38,631
I don't know what that is.

160
00:10:38,631 --> 00:10:43,741
And then of course, keep 2, and
the tunnel should come up.

161
00:10:43,741 --> 00:10:49,169
Router 1's side should come
up once the keeps kick in.

162
00:10:49,169 --> 00:10:50,566
I'm not going to leave
the keeps on there

163
00:10:50,566 --> 00:10:55,759
because as we move forward
to other interfaces,

164
00:10:55,759 --> 00:10:57,308
interface types,
it's not going to

165
00:10:57,308 --> 00:10:58,891
work because keepalives are only

166
00:10:58,891 --> 00:11:00,494
supported on GRE tunnels.

167
00:11:07,956 --> 00:11:09,063
In fact, it may not work without

168
00:11:09,063 --> 00:11:12,129
an IPv4 address
anyway since both

169
00:11:12,129 --> 00:11:15,556
sides just went down.

170
00:11:15,556 --> 00:11:17,460
We're not going to keep
them on there anyway.

171
00:11:17,460 --> 00:11:20,661
Just do no keeps. We're about to
change this all to GRE anyway.

172
00:11:25,692 --> 00:11:26,676
There you can tell I'm ready.

173
00:11:26,676 --> 00:11:30,054
It's working because EIGRP just
came up over the interface.

174
00:11:30,054 --> 00:11:37,106
Now if I said, do show
ipv6 route eigrp,

175
00:11:37,106 --> 00:11:39,928
now we have the
routes on both sides.

176
00:11:39,928 --> 00:11:42,902
If we were to go back to say,

177
00:11:42,902 --> 00:11:52,638
switch 1, do show
ipv6 route eigrp,

178
00:11:52,638 --> 00:11:58,995
and we were to say do
trace, and let's trace to--

179
00:12:02,421 --> 00:12:04,196
let's trace to what
would effectively

180
00:12:04,208 --> 00:12:05,854
be the client's
subnet over there.

181
00:12:08,149 --> 00:12:13,240
::8. There you can see that

182
00:12:13,240 --> 00:12:17,101
it hits the tunnel on router 1.

183
00:12:17,101 --> 00:12:19,401
Comes out the tunnel on the
other end on router 4,

184
00:12:19,401 --> 00:12:21,962
and gets delivered to switch 2.

185
00:12:21,962 --> 00:12:23,629
So very, very simple.

186
00:12:23,629 --> 00:12:25,599
Of course, the glory of

187
00:12:25,599 --> 00:12:26,860
this whole thing
is you go to say

188
00:12:26,860 --> 00:12:29,739
router 2, which is a
device in the core,

189
00:12:29,739 --> 00:12:34,192
and you say, do show
ipv6 interface brief,

190
00:12:34,192 --> 00:12:36,771
and he has nothing.

191
00:12:36,771 --> 00:12:38,741
Unassigned, unassigned,
unassigned.

192
00:12:38,741 --> 00:12:40,958
Everything's unassigned.

193
00:12:40,958 --> 00:12:43,693
Do show ipv6 route.

194
00:12:43,693 --> 00:12:46,887
He's not even running
IPv6 routing.

195
00:12:46,887 --> 00:12:52,443
So again, this is just
standard GRE tunnel.

196
00:12:52,443 --> 00:12:54,144
Now, if we take a look at this

197
00:12:54,144 --> 00:12:56,304
and we go over to-- let's go

198
00:12:56,304 --> 00:12:57,836
back to switch 1.

199
00:13:00,114 --> 00:13:03,546
And let's say, do ping.

200
00:13:03,546 --> 00:13:06,364
Protocol is IPv6.

201
00:13:08,864 --> 00:13:15,848
Target, we'll just ping his
loopback just to make it easy.

202
00:13:15,848 --> 00:13:18,202
Repeat count 1.

203
00:13:18,202 --> 00:13:19,982
Datagram size, I don't care.

204
00:13:19,982 --> 00:13:21,941
Time out, 1.

205
00:13:21,941 --> 00:13:23,559
This is all FastEthernet.

206
00:13:23,559 --> 00:13:25,545
It shouldn't take long.

207
00:13:25,545 --> 00:13:27,932
Extended commands, yes.

208
00:13:27,932 --> 00:13:30,033
Source interface, I don't care.

209
00:13:30,033 --> 00:13:31,995
UDP protocol is no.

210
00:13:31,995 --> 00:13:34,458
Verbos is no. I don't care about

211
00:13:34,458 --> 00:13:38,514
changing the precedence
or the DSCP value.

212
00:13:38,514 --> 00:13:40,902
I don't necessarily
need to see hop by hop.

213
00:13:40,902 --> 00:13:42,515
We did that with the trace.

214
00:13:42,515 --> 00:13:44,243
Destination, no.

215
00:13:44,243 --> 00:13:47,222
Sweep range of sizes, yes.

216
00:13:47,222 --> 00:13:51,780
Minimum sweep size, I'm
going to do 1401 bytes.

217
00:13:51,780 --> 00:13:54,354
I'm going to say the max size, I
don't care because I'm going to

218
00:13:54,354 --> 00:13:57,386
stop as soon as it
starts failing anyway.

219
00:13:57,386 --> 00:13:59,096
The sweep interval is 1.

220
00:13:59,096 --> 00:14:00,939
That's how much
it's going jump by.

221
00:14:00,939 --> 00:14:02,996
What this means, by the way,
if you've never done this,

222
00:14:02,996 --> 00:14:04,375
this means it's
going to start with

223
00:14:04,375 --> 00:14:06,995
a ping of 1401 and then it will

224
00:14:06,995 --> 00:14:11,135
do 1402, 1403, 1404, and so on.

225
00:14:11,135 --> 00:14:15,391
If I said sweep interval
2, it would do 1401,

226
00:14:15,391 --> 00:14:18,963
1403, 1405. It would jump by 2.

227
00:14:18,963 --> 00:14:21,152
That's how much it's
going to jump each time.

228
00:14:21,152 --> 00:14:23,076
I'm going to start
the ping and as

229
00:14:33,920 --> 00:14:33,933
soon as it hits--
Let's stop that.

230
00:14:33,933 --> 00:14:39,958
There we go. The problem is this
isn't really going to do what I

231
00:14:39,958 --> 00:14:42,670
want it to do, unfortunately.

232
00:14:42,670 --> 00:14:45,607
This has to do with that
whole packet too big thing.

233
00:14:45,607 --> 00:14:49,291
We could count those
packets right there

234
00:14:49,291 --> 00:14:50,587
to figure out what the actual

235
00:14:50,587 --> 00:14:55,552
MTU is, but, do show
interface tunnel 0.

236
00:14:59,143 --> 00:15:02,170
Unfortunately, because we

237
00:15:02,170 --> 00:15:04,986
don't have the
fragmentation built

238
00:15:04,986 --> 00:15:10,932
in to IPv6 like we talked about
back in the introduction part,

239
00:15:10,932 --> 00:15:12,870
unfortunately,
sweeping the range

240
00:15:12,870 --> 00:15:15,352
of sizes isn't quite as nice as

241
00:15:15,352 --> 00:15:20,043
it is on IPv4 because you don't
have a don't fragment bit.

242
00:15:20,043 --> 00:15:23,486
The devices in the middle
can't fragment anyway.

243
00:15:23,486 --> 00:15:27,126
Basically what it
means is that at

244
00:15:27,126 --> 00:15:32,670
1478 should be what we managed

245
00:15:32,670 --> 00:15:36,983
to get through here
from a payload size,

246
00:15:36,983 --> 00:15:40,769
but that's not entirely
correct either.

247
00:15:40,769 --> 00:15:42,240
Again, it's not going
to be a really

248
00:15:42,240 --> 00:15:44,352
good indicator here as to what

249
00:15:44,352 --> 00:15:45,877
our actual MTU is.

250
00:15:48,863 --> 00:15:51,611
Why did that not show me that?

251
00:15:51,611 --> 00:15:52,933
All on the switch.

252
00:16:02,120 --> 00:16:02,771
You can see

253
00:16:02,771 --> 00:16:05,559
that the way IPv6
handles the MTU,

254
00:16:05,559 --> 00:16:07,196
is completely different
anyway because

255
00:16:07,196 --> 00:16:10,690
it uses the path MTU discovery.

256
00:16:10,690 --> 00:16:12,292
But basically, at
the end of the day,

257
00:16:12,292 --> 00:16:15,463
what we're saying here
is that right now,

258
00:16:15,463 --> 00:16:18,006
our tunnel and
protocol transport

259
00:16:18,006 --> 00:16:21,723
is GRE over IP,
which means we're

260
00:16:21,723 --> 00:16:26,643
getting an extra header
of GRE right now.

261
00:16:26,643 --> 00:16:27,664
The only thing we're running

262
00:16:27,664 --> 00:16:30,647
over this tunnel is IPv6.

263
00:16:30,647 --> 00:16:36,287
So there's actually no need
whatsoever to have a GRE header.

264
00:16:36,287 --> 00:16:37,455
But that's the default.

265
00:16:37,455 --> 00:16:39,730
That's it. It's very simple.

266
00:16:39,730 --> 00:16:41,746
The advantage to
this, of course,

267
00:16:41,746 --> 00:16:48,123
is that we could
actually run IPv4

268
00:16:48,123 --> 00:16:52,063
over this if we
wanted to and so on.

269
00:16:52,063 --> 00:16:53,612
We're going to change
that in a minute.

270
00:16:53,612 --> 00:16:54,966
But I just wanted you
to know right now

271
00:16:54,966 --> 00:16:56,082
that the tunnel transport

272
00:16:56,082 --> 00:17:00,230
MTU is 1476. So
we'll come back and

273
00:17:00,230 --> 00:17:02,282
take a look at this tunnel again

274
00:17:02,282 --> 00:17:04,335
in our next section.

275
00:17:04,335 --> 00:17:06,137
What we'll do in the
next section is

276
00:17:06,137 --> 00:17:09,942
we will get rid
of the GRE header

277
00:17:09,942 --> 00:17:13,550
and start moving into
other tunnel types.

278
00:17:13,550 --> 00:17:18,971
[music]