1
00:00:00,000 --> 00:00:00,982

2
00:00:00,982 --> 00:00:04,419
[AUDIO LOGO]

3
00:00:04,419 --> 00:00:08,060

4
00:00:08,060 --> 00:00:13,530
So we just saw how to do BGP
filtering to filter things out

5
00:00:13,530 --> 00:00:16,800
based on maybe their
prefix or their AS path.

6
00:00:16,800 --> 00:00:20,760
But sometimes it's not
about what you're getting,

7
00:00:20,760 --> 00:00:23,880
but how much of
something you're getting.

8
00:00:23,880 --> 00:00:26,820
If we're getting
too many BGP routes,

9
00:00:26,820 --> 00:00:30,280
that can definitely
overwhelm our routers.

10
00:00:30,280 --> 00:00:34,080
So BGP route
limiting can be used

11
00:00:34,080 --> 00:00:37,020
to ensure that we don't
get too many prefixes

12
00:00:37,020 --> 00:00:38,460
from our neighbors.

13
00:00:38,460 --> 00:00:42,450
Perfect example would
be on the customer side.

14
00:00:42,450 --> 00:00:46,800
Maybe again, they asked for
only the default route or maybe

15
00:00:46,800 --> 00:00:50,520
partial routes, and the service
provider instead sends them

16
00:00:50,520 --> 00:00:52,590
the entire BGP table.

17
00:00:52,590 --> 00:00:55,380
Their router may not
be able to handle that.

18
00:00:55,380 --> 00:00:58,650
It does take a large
amount of RAM and CPU

19
00:00:58,650 --> 00:01:01,420
to handle the internet
routing table appropriately.

20
00:01:01,420 --> 00:01:04,739
So we can limit
how many prefixes

21
00:01:04,739 --> 00:01:07,990
we'll accept from a
neighbor, and that's

22
00:01:07,990 --> 00:01:10,100
the premise of this video.

23
00:01:10,100 --> 00:01:13,340
So let's start by taking a
look at some of the details.

24
00:01:13,340 --> 00:01:16,030
And then, of course, we'll
jump into the command-line

25
00:01:16,030 --> 00:01:20,210
and see this feature in use.

26
00:01:20,210 --> 00:01:22,780
So the idea behind
route limiting

27
00:01:22,780 --> 00:01:27,850
is to protect the router
from being overwhelmed

28
00:01:27,850 --> 00:01:30,160
with too many prefixes.

29
00:01:30,160 --> 00:01:33,520
Again, a good example would
be the internet routing table,

30
00:01:33,520 --> 00:01:35,680
or maybe there's
a bigger mistake,

31
00:01:35,680 --> 00:01:39,460
and somebody's sending
multiple copies of the table.

32
00:01:39,460 --> 00:01:41,890
And again, this would
come down to the idea

33
00:01:41,890 --> 00:01:46,600
that even a legitimate neighbor
that I normally accept routes

34
00:01:46,600 --> 00:01:50,560
from possibly got
misconfigured, or the filtering

35
00:01:50,560 --> 00:01:53,390
is not configured properly
on it, to begin with.

36
00:01:53,390 --> 00:01:55,840
So there's a couple
reasons why we could

37
00:01:55,840 --> 00:01:58,240
get too many prefixes here.

38
00:01:58,240 --> 00:02:04,300
The real focus is, this is the
amount of the routes received,

39
00:02:04,300 --> 00:02:07,060
not the actual prefixes.

40
00:02:07,060 --> 00:02:10,960
Again, actual prefixes would
come down to prefix filtering.

41
00:02:10,960 --> 00:02:14,050
This is all about
how many routes.

42
00:02:14,050 --> 00:02:17,950
Now, as we're configuring
this, an important design point

43
00:02:17,950 --> 00:02:22,410
here is, we do need
to allow for growth.

44
00:02:22,410 --> 00:02:24,602
And when we see some
recommendations in a moment,

45
00:02:24,602 --> 00:02:26,060
we'll see that
we're actually going

46
00:02:26,060 --> 00:02:28,730
to allow for a lot of growth.

47
00:02:28,730 --> 00:02:31,880
We're not going to want to
have to go change this feature.

48
00:02:31,880 --> 00:02:34,860
Every time, things
change just a little bit.

49
00:02:34,860 --> 00:02:37,790
So for example,
let's say a customer

50
00:02:37,790 --> 00:02:40,700
is sending us four prefixes.

51
00:02:40,700 --> 00:02:43,850
I wouldn't want to
set this to 5 or 7.

52
00:02:43,850 --> 00:02:47,840
I would probably set it
to 30 or maybe even 40.

53
00:02:47,840 --> 00:02:50,810
That way, they have
plenty of room to grow.

54
00:02:50,810 --> 00:02:54,740
The idea here is, I don't
want them sending me 900,000.

55
00:02:54,740 --> 00:02:57,020
I don't want them
to accidentally feed

56
00:02:57,020 --> 00:02:59,450
the entire internet
routing table back to me

57
00:02:59,450 --> 00:03:00,990
from a customer site.

58
00:03:00,990 --> 00:03:04,790
So again, we don't have to
be real close to the number

59
00:03:04,790 --> 00:03:07,050
that they're actually
trying to send.

60
00:03:07,050 --> 00:03:09,142
So we'll, again, see
some examples of this,

61
00:03:09,142 --> 00:03:10,850
but let's just keep
that in mind as we're

62
00:03:10,850 --> 00:03:13,200
looking over how this works.

63
00:03:13,200 --> 00:03:15,420
And I did just drop this number.

64
00:03:15,420 --> 00:03:17,700
But as of the time
of this recording,

65
00:03:17,700 --> 00:03:21,390
I just looked at one of AT&T's
looking glass routers earlier,

66
00:03:21,390 --> 00:03:23,760
and the current
routing table is just

67
00:03:23,760 --> 00:03:30,040
above 900,000 prefixes
for the internet.

68
00:03:30,040 --> 00:03:33,300
So looking at route limiting
from the service provider's

69
00:03:33,300 --> 00:03:38,640
perspective, again, the idea
is, let's prevent the customer

70
00:03:38,640 --> 00:03:41,160
from sending too many prefixes.

71
00:03:41,160 --> 00:03:43,800
But again, like I
said a moment ago,

72
00:03:43,800 --> 00:03:46,200
we're not going to
go real close here.

73
00:03:46,200 --> 00:03:51,330
We're going to go something like
10 times the amount of prefixes

74
00:03:51,330 --> 00:03:53,350
that we're expecting.

75
00:03:53,350 --> 00:03:56,670
So if we're expecting them
to send us four prefixes,

76
00:03:56,670 --> 00:03:58,570
let's set this to 40.

77
00:03:58,570 --> 00:04:02,310
We're just trying to make sure
they're not sending us 100,000.

78
00:04:02,310 --> 00:04:04,830
We can also use this
as a service provider

79
00:04:04,830 --> 00:04:08,880
to make sure that other service
providers don't make a mistake,

80
00:04:08,880 --> 00:04:11,820
and they don't try to
send us too many prefixes.

81
00:04:11,820 --> 00:04:14,190
Again, given the current
size of the internet

82
00:04:14,190 --> 00:04:16,589
routing table,
probably a good idea

83
00:04:16,589 --> 00:04:19,500
to set this to about 2 million.

84
00:04:19,500 --> 00:04:24,480
That way again, they can't send
us three, four, seven copies

85
00:04:24,480 --> 00:04:26,370
of the internet routing table.

86
00:04:26,370 --> 00:04:28,810
We only want one copy.

87
00:04:28,810 --> 00:04:31,820
2 million should give us
plenty of room to grow,

88
00:04:31,820 --> 00:04:35,000
so we don't have to come
change this all the time.

89
00:04:35,000 --> 00:04:38,990
Now, looking at this from
the customers' perspective,

90
00:04:38,990 --> 00:04:43,540
it really depends on what routes
they've requested to receive.

91
00:04:43,540 --> 00:04:48,070
If they've requested to receive
only the default route, well,

92
00:04:48,070 --> 00:04:52,240
then, maybe they would want
to set this to say, 2 to 5,

93
00:04:52,240 --> 00:04:54,370
and they may never
grow since they're

94
00:04:54,370 --> 00:04:57,410
just asking for the
default. But again,

95
00:04:57,410 --> 00:05:00,940
giving a little bit of
overhead, not a bad idea here.

96
00:05:00,940 --> 00:05:05,050
If they ask for partial
updates, once again,

97
00:05:05,050 --> 00:05:09,160
it should be right around 10
times the service providers'

98
00:05:09,160 --> 00:05:10,400
prefixes.

99
00:05:10,400 --> 00:05:12,880
Now, how would they know this?

100
00:05:12,880 --> 00:05:15,100
Service providers
may publish it,

101
00:05:15,100 --> 00:05:18,520
or they could call in
to Customer Support

102
00:05:18,520 --> 00:05:23,140
and get this number, basically
just ask how many prefixes

103
00:05:23,140 --> 00:05:25,000
are in the partial table.

104
00:05:25,000 --> 00:05:28,615
Most providers have that
information available.

105
00:05:28,615 --> 00:05:30,820
Worst-case scenario,
of course, they

106
00:05:30,820 --> 00:05:32,590
could just allow
all the prefixes,

107
00:05:32,590 --> 00:05:34,880
and then they could count them.

108
00:05:34,880 --> 00:05:36,550
I don't mean
literally count them.

109
00:05:36,550 --> 00:05:38,890
I just mean, look
at the BGP table.

110
00:05:38,890 --> 00:05:41,560
And, of course, just like
another service provider,

111
00:05:41,560 --> 00:05:43,510
if they're getting
full internet routes,

112
00:05:43,510 --> 00:05:46,420
then we probably want to just
go ahead with the 2 million

113
00:05:46,420 --> 00:05:48,220
to give room to grow.

114
00:05:48,220 --> 00:05:54,670
Now, this feature is enabled
on IOS XR by default,

115
00:05:54,670 --> 00:05:58,060
and it has some pre-built
default settings.

116
00:05:58,060 --> 00:06:04,750
It allows 512,000
prefixes for IPv4 unicast,

117
00:06:04,750 --> 00:06:13,990
128,000 for IPv4 multicast,
128,000 for IPv6 unicast,

118
00:06:13,990 --> 00:06:20,770
and 512,000 for VPNv4 unicast.

119
00:06:20,770 --> 00:06:23,290
Now, chances are,
these numbers probably

120
00:06:23,290 --> 00:06:26,200
will not meet the criteria
of your environment.

121
00:06:26,200 --> 00:06:28,450
But of course, it's
a place to start,

122
00:06:28,450 --> 00:06:31,820
and we can always adjust
these numbers, as needed.

123
00:06:31,820 --> 00:06:33,850
Also, just a quick
note down here.

124
00:06:33,850 --> 00:06:38,260
It's not really related to
route limiting but just for lack

125
00:06:38,260 --> 00:06:39,970
of a better place to put this.

126
00:06:39,970 --> 00:06:43,330
There's also a
default maximum number

127
00:06:43,330 --> 00:06:49,460
of neighbors set to 4000
by default. Once again,

128
00:06:49,460 --> 00:06:52,580
this can be changed but
just be aware of it.

129
00:06:52,580 --> 00:06:54,810
So let's jump into
the command-line.

130
00:06:54,810 --> 00:06:57,770
We'll take a look at
how this feature works.

131
00:06:57,770 --> 00:07:00,202
Let's bring up our
topology first and talk

132
00:07:00,202 --> 00:07:01,910
about what we're going
to be configuring,

133
00:07:01,910 --> 00:07:04,320
and then we'll jump in
and take a look at it.

134
00:07:04,320 --> 00:07:08,930
So we do want to look at this
feature on both IOS XR and IOS

135
00:07:08,930 --> 00:07:09,860
XE.

136
00:07:09,860 --> 00:07:13,520
So for IOS XR, we'll be
focusing on router 1.

137
00:07:13,520 --> 00:07:18,440
And for iOS XE, we'll
be focusing on router 7.

138
00:07:18,440 --> 00:07:21,120
And what we're going
to do on router 10,

139
00:07:21,120 --> 00:07:25,370
we're going to have it generate
a whole bunch of prefixes

140
00:07:25,370 --> 00:07:27,920
and send them into BGP.

141
00:07:27,920 --> 00:07:32,310
And we'll see how this limiting
works on our different routers.

142
00:07:32,310 --> 00:07:34,260
So let's jump into
the command-line.

143
00:07:34,260 --> 00:07:37,820
We'll start on router 7
just to see what's already

144
00:07:37,820 --> 00:07:41,490
going on in our BGP table.

145
00:07:41,490 --> 00:07:43,910
So the first thing we'll
note is if we do a show ip

146
00:07:43,910 --> 00:07:48,350
bgp on router 7, we
don't have any prefixes

147
00:07:48,350 --> 00:07:50,270
coming from router 10.

148
00:07:50,270 --> 00:07:54,380
This is quite simply because
they are both in the same AS.

149
00:07:54,380 --> 00:07:58,460
So just to get the prefixes to
be accepted here on router 7,

150
00:07:58,460 --> 00:08:02,225
we're just going to go ahead
and allowas-in on router 7.

151
00:08:02,225 --> 00:08:05,620

152
00:08:05,620 --> 00:08:09,370
Let's do our show ip bgp again,
and we see that now, we're

153
00:08:09,370 --> 00:08:13,180
getting prefixes coming
from router 10, of course,

154
00:08:13,180 --> 00:08:18,160
through router 6, specifically
the ones coming from 65001.

155
00:08:18,160 --> 00:08:21,340
If we do a show
ip bgp summary, we

156
00:08:21,340 --> 00:08:24,400
can see specifically
that from router 6,

157
00:08:24,400 --> 00:08:27,413
we are currently
getting eight prefixes.

158
00:08:27,413 --> 00:08:29,830
What we're going to do is we're
going to go ahead and make

159
00:08:29,830 --> 00:08:33,010
that a nice even
80, which means,

160
00:08:33,010 --> 00:08:38,110
we're going to make 72 loopback
interfaces on router 10

161
00:08:38,110 --> 00:08:40,909
and bring them all into BGP.

162
00:08:40,909 --> 00:08:43,120
So let's jump over to router 10.

163
00:08:43,120 --> 00:08:47,900
We'll go into configuration
mode, and we'll get typing.

164
00:08:47,900 --> 00:08:49,370
No, s just kidding.

165
00:08:49,370 --> 00:08:51,920
We're going to use an
EEM script for this,

166
00:08:51,920 --> 00:08:55,730
but we are going to
create the 72 loopbacks.

167
00:08:55,730 --> 00:08:58,370
So we'll just
paste my script in.

168
00:08:58,370 --> 00:09:00,650
If you want to look
over this, it's

169
00:09:00,650 --> 00:09:04,790
creating a variable called
L. It's starting in at 1.

170
00:09:04,790 --> 00:09:09,650
It's going to increment up to
72, make a loopback interface,

171
00:09:09,650 --> 00:09:16,940
and assign the subnet 10.100
dot whatever number it is dot 1

172
00:09:16,940 --> 00:09:19,070
with a /24.

173
00:09:19,070 --> 00:09:22,280
So since this event
is set to none,

174
00:09:22,280 --> 00:09:25,280
we'll just go ahead
and run it manually.

175
00:09:25,280 --> 00:09:28,130
If you look up top, I
just called it LOOP1.

176
00:09:28,130 --> 00:09:31,460
Just as a note, I do have
the debug running here

177
00:09:31,460 --> 00:09:34,490
for the CLI output
of the EEM script

178
00:09:34,490 --> 00:09:36,620
just so we can see it happening.

179
00:09:36,620 --> 00:09:39,120
And in case you try to
run this on your own,

180
00:09:39,120 --> 00:09:42,260
also note that I have
sped this up in the video.

181
00:09:42,260 --> 00:09:45,840
It actually takes quite a
few seconds for this to run.

182
00:09:45,840 --> 00:09:48,810
I also really like doing
the debug because I just

183
00:09:48,810 --> 00:09:53,610
find this line rather
humorous that tty is now going

184
00:09:53,610 --> 00:09:55,500
through its death sequence.

185
00:09:55,500 --> 00:09:57,630
All that really
means is the router

186
00:09:57,630 --> 00:10:02,650
is killing the CLI instance that
it was using to run the script.

187
00:10:02,650 --> 00:10:06,510
So we can see that all 72
interfaces have come up.

188
00:10:06,510 --> 00:10:08,310
If we do a show ip
interface brief,

189
00:10:08,310 --> 00:10:09,930
we can see all these
loopbacks, and we

190
00:10:09,930 --> 00:10:13,330
see that the third octet matches
up with the loopback number.

191
00:10:13,330 --> 00:10:16,450
So this should be good to go.

192
00:10:16,450 --> 00:10:19,290
If we do a show ip
bgp, all of these

193
00:10:19,290 --> 00:10:21,880
are now being injected
into BGP as well.

194
00:10:21,880 --> 00:10:24,630
So now, let's go back
over to router 7.

195
00:10:24,630 --> 00:10:26,607
We don't need to see
all these prefixes.

196
00:10:26,607 --> 00:10:27,690
We'll just do the summary.

197
00:10:27,690 --> 00:10:30,870
Again, it's about the number
of prefixes, not what they are.

198
00:10:30,870 --> 00:10:36,040
And we see we now have a nice,
even 80 prefixes coming in.

199
00:10:36,040 --> 00:10:38,070
So now, let's
implement our limit.

200
00:10:38,070 --> 00:10:43,040
We'll go into configuration
mode in the router BGP,

201
00:10:43,040 --> 00:10:47,150
and we will say that for
neighbor 50.0.167.6, which

202
00:10:47,150 --> 00:10:51,320
is router 6, we're going
to set the maximum prefix.

203
00:10:51,320 --> 00:10:55,850
It wants to know the maximum
number, let's say, 100.

204
00:10:55,850 --> 00:10:59,300
Then it's going to ask
what the threshold is.

205
00:10:59,300 --> 00:11:01,250
We're going to say 90.

206
00:11:01,250 --> 00:11:04,100
And then we have
two options left.

207
00:11:04,100 --> 00:11:10,170
We can either restart, or
we can do a warning-only.

208
00:11:10,170 --> 00:11:13,920
If we do warning-only
here, it won't actually

209
00:11:13,920 --> 00:11:15,660
take down the neighborship.

210
00:11:15,660 --> 00:11:19,440
The default behavior is to
take down the neighborship,

211
00:11:19,440 --> 00:11:22,090
and that's what we're
going to do here.

212
00:11:22,090 --> 00:11:24,300
So we're going to
tell it to restart,

213
00:11:24,300 --> 00:11:26,860
which, you don't have to
configure this either,

214
00:11:26,860 --> 00:11:30,310
but we'll tell it to restart
after, say, 2 minutes.

215
00:11:30,310 --> 00:11:32,640
So let's go ahead and hit Enter.

216
00:11:32,640 --> 00:11:35,110
We'll exit out of configuration.

217
00:11:35,110 --> 00:11:39,420
And let's go configure the
same settings on router 1.

218
00:11:39,420 --> 00:11:42,570
So router 1 will go
to Configuration mode.

219
00:11:42,570 --> 00:11:45,390
This is router bgp 100.

220
00:11:45,390 --> 00:11:50,640
Same neighbor address for
the address family IPv4.

221
00:11:50,640 --> 00:11:52,440
We'll set the maximum prefix.

222
00:11:52,440 --> 00:11:56,160
And as we can see once
again, maximum number.

223
00:11:56,160 --> 00:11:58,110
We pretty much have
the same options

224
00:11:58,110 --> 00:12:01,290
except for the discard
extra path option.

225
00:12:01,290 --> 00:12:04,360
We'll just set the
threshold to 90.

226
00:12:04,360 --> 00:12:07,290
And again, we'll
restart after 2 minutes.

227
00:12:07,290 --> 00:12:10,790
Notice, warning-only is
an option here as well.

228
00:12:10,790 --> 00:12:15,110
Let's commit our changes
and exit configuration.

229
00:12:15,110 --> 00:12:17,890
So as it is right
now, we're getting

230
00:12:17,890 --> 00:12:22,280
80 prefixes, which is, of
course, 10 below our threshold.

231
00:12:22,280 --> 00:12:24,700
So let's go back
over to router 10

232
00:12:24,700 --> 00:12:29,680
and create 15 more loopbacks
to push us over the threshold

233
00:12:29,680 --> 00:12:32,630
but not reach the
limit just yet.

234
00:12:32,630 --> 00:12:36,000
Of course, we have
another script for that.

235
00:12:36,000 --> 00:12:37,740
These are going
to be odd numbers

236
00:12:37,740 --> 00:12:43,740
because it's going to create
loopbacks 73 through 87.

237
00:12:43,740 --> 00:12:47,250
So the numbers of the loopbacks
themselves are going to be odd.

238
00:12:47,250 --> 00:12:50,910
However, it's going to create
15 additional loopbacks.

239
00:12:50,910 --> 00:12:53,430
So we can see that
that's configured.

240
00:12:53,430 --> 00:12:57,660
And if we go check our
routers, router 7, we see here,

241
00:12:57,660 --> 00:12:59,490
we already have a
warning, telling us

242
00:12:59,490 --> 00:13:01,620
that the number of
prefixes received

243
00:13:01,620 --> 00:13:06,310
has reached 91 towards
our max of 100.

244
00:13:06,310 --> 00:13:09,610
Now, if we do a
show ip bgp summary,

245
00:13:09,610 --> 00:13:11,940
we can see that
it's actually 95.

246
00:13:11,940 --> 00:13:14,580
We said we were
adding 15, but we just

247
00:13:14,580 --> 00:13:18,420
got the one message when it
actually broke the threshold.

248
00:13:18,420 --> 00:13:20,940
Let's see what
router 1 is saying.

249
00:13:20,940 --> 00:13:24,210
Router 1, we can see it's
almost the exact same message

250
00:13:24,210 --> 00:13:28,500
that we've now reached 91
towards the max of 100.

251
00:13:28,500 --> 00:13:31,450
We've broken the threshold.

252
00:13:31,450 --> 00:13:33,300
So you know what
we have to do now.

253
00:13:33,300 --> 00:13:36,000
We have to add more
loopbacks on router 10

254
00:13:36,000 --> 00:13:39,330
to push it over the maximum
and make these routers actually

255
00:13:39,330 --> 00:13:42,050
shut down their neighborship.

256
00:13:42,050 --> 00:13:44,663
So we'll just add some
more loopbacks here.

257
00:13:44,663 --> 00:13:46,580
We don't need to watch
it run this whole time.

258
00:13:46,580 --> 00:13:48,090
We'll just jump ahead.

259
00:13:48,090 --> 00:13:51,590
So now, router 10 has
created more loopbacks.

260
00:13:51,590 --> 00:13:54,110
Now, let's go back
over to router 7.

261
00:13:54,110 --> 00:13:57,050
We see that so far, it's just
given us another warning.

262
00:13:57,050 --> 00:14:00,800
And now, we see that it
actually goes over the maximum

263
00:14:00,800 --> 00:14:03,320
and resets the peer.

264
00:14:03,320 --> 00:14:05,750
Let's go take a
look at router 1.

265
00:14:05,750 --> 00:14:08,870
We can see that
router 1 here has also

266
00:14:08,870 --> 00:14:14,130
hit the maximum prefixes
and shut down the neighbor.

267
00:14:14,130 --> 00:14:18,810
Now, we configure these
devices to recover or restart

268
00:14:18,810 --> 00:14:22,200
the connection after 2
minutes, but in order

269
00:14:22,200 --> 00:14:25,590
to do that, we would have
to get the prefixes back

270
00:14:25,590 --> 00:14:27,120
below the maximum.

271
00:14:27,120 --> 00:14:29,070
To do that, we're
just going to go

272
00:14:29,070 --> 00:14:30,990
undo everything we
did on router 10

273
00:14:30,990 --> 00:14:34,800
and just take out all
those new loopbacks.

274
00:14:34,800 --> 00:14:38,100
On router 10, we're simply going
to do a configuration rollback

275
00:14:38,100 --> 00:14:42,480
to the saved configuration that
I have from before we started

276
00:14:42,480 --> 00:14:43,780
this lab.

277
00:14:43,780 --> 00:14:47,640
And here, we can see it taking
down all of those interfaces.

278
00:14:47,640 --> 00:14:49,660
If we do a show ip
interface brief,

279
00:14:49,660 --> 00:14:51,540
we see we're back to just the--

280
00:14:51,540 --> 00:14:53,610
we'll call them the
normal interfaces

281
00:14:53,610 --> 00:14:57,700
that we normally have on router
10 in our lab environment here.

282
00:14:57,700 --> 00:15:00,180
This means that
router 1 and router 7

283
00:15:00,180 --> 00:15:02,640
should actually recover
their neighborships

284
00:15:02,640 --> 00:15:04,690
when they try again.

285
00:15:04,690 --> 00:15:09,480
So it's been our 2 minutes here,
and we can see here on router 7

286
00:15:09,480 --> 00:15:13,570
that the neighborship with
router 6 has come back up.

287
00:15:13,570 --> 00:15:16,530
And if we look at router
1, we see that router 6

288
00:15:16,530 --> 00:15:18,750
is back up here as well.

289
00:15:18,750 --> 00:15:22,500
We look at our summary, we
can see that here on router 1,

290
00:15:22,500 --> 00:15:27,060
it's back to just learning those
eight prefixes from router 6

291
00:15:27,060 --> 00:15:30,540
as it was when we started.

292
00:15:30,540 --> 00:15:34,490
So in this video, we
took a look at BGP route

293
00:15:34,490 --> 00:15:38,180
limiting, where we can
limit the number of routes

294
00:15:38,180 --> 00:15:41,900
that a router will receive
or accept from a neighbor

295
00:15:41,900 --> 00:15:44,750
before it gives us
first a warning when

296
00:15:44,750 --> 00:15:46,220
we get to the threshold.

297
00:15:46,220 --> 00:15:48,650
And then when it
gets to the maximum,

298
00:15:48,650 --> 00:15:51,980
well, we saw it will either shut
down the neighborship, which

299
00:15:51,980 --> 00:15:54,770
is what it does by default, by
the way, if we don't configure

300
00:15:54,770 --> 00:16:00,080
the other options, or we can
use that option warning-only.

301
00:16:00,080 --> 00:16:02,090
In which case, it
will continually

302
00:16:02,090 --> 00:16:04,760
yell at us, of course,
but it will not actually

303
00:16:04,760 --> 00:16:07,190
shut down the BGP neighborship.

304
00:16:07,190 --> 00:16:09,590
We also saw that
there's an option

305
00:16:09,590 --> 00:16:13,040
there to restart after a
certain number of minutes

306
00:16:13,040 --> 00:16:15,170
we chose to do to.

307
00:16:15,170 --> 00:16:17,480
I hope this has been
informative for you,

308
00:16:17,480 --> 00:16:20,710
and I'd like to thank
you for viewing.

309
00:16:20,710 --> 00:16:24,000