WEBVTT

1
00:00.000 --> 00:06.360
First, top redundancy protocols are a critical part of adding redundancy to our

2
00:06.360 --> 00:07.860
network.

3
00:07.860 --> 00:13.060
This is specifically going to help us out with our default gateways.

4
00:13.060 --> 00:16.260
The fault gateways are how we get off our subnet.

5
00:16.260 --> 00:23.760
If we have a user down here and they're part of the 10.100.1.0/24 subnet, let's

6
00:23.760 --> 00:23.960
say they

7
00:23.960 --> 00:29.180
have an IP address of .100, then they're going to need to get to other subnets

8
00:29.180 --> 00:29.620
via the

9
00:29.620 --> 00:30.620
default gateway.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

9
00:29.620 --> 00:30.620
default gateway.

10
00:30.620 --> 00:35.380
Now, these days, in our modern networks, we're going to do that with multi-lay

11
00:35.380 --> 00:36.340
er switches.

12
00:36.340 --> 00:37.900
We're going to have an SVI here.

13
00:37.900 --> 00:41.900
That'd be a switched virtual interface or a VLAN interface.

14
00:41.900 --> 00:45.060
Let's say that this is VLAN 100.

15
00:45.060 --> 00:48.380
We're going to call this layer to segment VLAN 100.

16
00:48.380 --> 00:52.100
We create an SVI for VLAN 100.

17
00:52.100 --> 00:55.900
That'd be the interface of VLAN 100 configuration command.

18
00:55.900 --> 01:00.660
We add an IP address to it.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

18
00:55.900 --> 01:00.660
We add an IP address to it.

19
01:00.660 --> 01:07.820
As our default gateway, this device knows that it's default gateway is .254.

20
01:07.820 --> 01:10.580
It can get that via DHCP.

21
01:10.580 --> 01:12.700
It knows how to get off the subnet.

22
01:12.700 --> 01:16.340
Now, we have a way to route to other networks.

23
01:16.340 --> 01:19.500
Let's say that I have another SVI here.

24
01:19.500 --> 01:22.140
This is for VLAN 200.

25
01:22.140 --> 01:28.140
This has its own network segment with another endpoint over here.

26
01:28.140 --> 01:31.260
Let's just actually say that this is a printer.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

26
01:28.140 --> 01:31.260
Let's just actually say that this is a printer.

27
01:31.260 --> 01:35.140
We're trying to print something off and it's on VLAN 200.

28
01:35.140 --> 01:45.340
This is 10.200.0/24, and then we'll add an IP address here of .50.

29
01:45.340 --> 01:46.340
How do we get there?

30
01:46.340 --> 01:52.140
This host is going to perform an ARP or the default gateway.

31
01:52.140 --> 01:56.710
This is going to matter for how we accomplish everything with first-hopper-d

32
01:56.710 --> 01:58.060
ensity protocols.

33
01:58.060 --> 02:00.180
ARP is going to play a big role.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

33
01:58.060 --> 02:00.180
ARP is going to play a big role.

34
02:00.180 --> 02:02.500
Either way, we're going to ARP.254.

35
02:02.500 --> 02:07.140
We're going to find it here on our yellow subnet.

36
02:07.140 --> 02:09.540
This SVI is part of the subnet.

37
02:09.540 --> 02:11.060
It can respond at layer 2.

38
02:11.060 --> 02:13.060
It says, "Hey, here I am."

39
02:13.060 --> 02:20.540
Then our device is going to send a packet over to the printer.

40
02:20.540 --> 02:26.880
The way it really works is it gets forwarded up layer 2 to the SVI and VLAN 100

41
02:26.880 --> 02:26.940
.

42
02:26.940 --> 02:31.650
We perform a routing operation internally to get from SVI or maybe able to say
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

42
02:26.940 --> 02:31.650
We perform a routing operation internally to get from SVI or maybe able to say

43
02:31.650 --> 02:32.700
VLAN 100

44
02:32.700 --> 02:35.580
interface to the VLAN 200 interface.

45
02:35.580 --> 02:39.940
We can perform a local forwarding operation down to the printer.

46
02:39.940 --> 02:43.860
The reverse traffic flow would look very similar.

47
02:43.860 --> 02:47.860
The default gateway is naturally a very important part of our network.

48
02:47.860 --> 02:52.630
If it were to go down for any reason, because network failures do occur, well,

49
02:52.630 --> 02:53.700
at this point,

50
02:53.700 --> 02:58.860
we would not be able to communicate between our different subnets.

51
02:58.860 --> 03:03.940
The redundancy of a gateway is going to be mission critical.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

51
02:58.860 --> 03:03.940
The redundancy of a gateway is going to be mission critical.

52
03:03.940 --> 03:07.700
The network industry could have evolved in different ways.

53
03:07.700 --> 03:13.300
We could have just had another gateway over here and it's maybe .253.

54
03:13.300 --> 03:18.220
Maybe we could have added multiple default gateways down here to the host.

55
03:18.220 --> 03:19.820
That's not how we did it.

56
03:19.820 --> 03:24.540
Instead, we said, "We're going to lock in on a single default gateway."

57
03:24.540 --> 03:29.420
We're going to rely on the network to perform the redundancy.

58
03:29.420 --> 03:32.060
How exactly do we go about doing this?
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

58
03:29.420 --> 03:32.060
How exactly do we go about doing this?

59
03:32.060 --> 03:37.660
Again, this is where first hop redundancy protocols come into play.

60
03:37.660 --> 03:40.900
If we break down the name, we're talking about a first hop in here.

61
03:40.900 --> 03:42.900
That's what the FH stands for.

62
03:42.900 --> 03:48.180
This would be our first hop, our default gateway, to get off of the network.

63
03:48.180 --> 03:51.660
Then redundancy protocol probably speaks for itself.

64
03:51.660 --> 03:56.580
We need to have a redundant default gateway out there on the network.

65
03:56.580 --> 03:57.900
How do we do this?

66
03:57.900 --> 04:03.620
We know that the default gateway is locked to .254.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

66
03:57.900 --> 04:03.620
We know that the default gateway is locked to .254.

67
04:03.620 --> 04:13.020
What we want to do is we want to create the concept of a virtual IP address.

68
04:13.020 --> 04:17.220
This virtual IP address is going to be shared between our nodes.

69
04:17.220 --> 04:23.240
.254 is actually going to become sort of this IP address that either one of

70
04:23.240 --> 04:24.780
these devices

71
04:24.780 --> 04:27.860
is capable of running at the same time.

72
04:27.860 --> 04:31.180
Now if we're using this as a virtual IP address, we probably want to change the
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

72
04:27.860 --> 04:31.180
Now if we're using this as a virtual IP address, we probably want to change the

73
04:31.180 --> 04:31.700
physical

74
04:31.700 --> 04:37.260
address of the first device to let's say just be .252.

75
04:37.260 --> 04:39.580
Can I ping .252?

76
04:39.580 --> 04:41.180
Can I ping .253?

77
04:41.180 --> 04:44.140
The answer is yes, I can ping both of those.

78
04:44.140 --> 04:48.020
They're both responding to those IP addresses.

79
04:48.020 --> 04:53.620
By the way, that device on the right would be also adding this to our VLAN 100

80
04:53.620 --> 04:54.580
interface.

81
04:54.580 --> 04:56.420
Just to be clear.

82
04:56.420 --> 05:01.380
Those IP addresses are active, but so is the virtual IP address.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

82
04:56.420 --> 05:01.380
Those IP addresses are active, but so is the virtual IP address.

83
05:01.380 --> 05:07.220
It's going to be managed by the concept of the active node.

84
05:07.220 --> 05:14.040
The active node runs the virtual IP address until it can't anymore until it

85
05:14.040 --> 05:15.620
goes offline.

86
05:15.620 --> 05:20.220
Then we introduce the concept of the failover.

87
05:20.220 --> 05:26.020
The failover node will run if the active node goes down.

88
05:26.020 --> 05:31.960
The active node goes down and then the failover node manages the virtual IP
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

88
05:26.020 --> 05:31.960
The active node goes down and then the failover node manages the virtual IP

89
05:31.960 --> 05:32.300
address.

90
05:32.300 --> 05:36.500
Importantly, we can't just survive on a virtual IP address.

91
05:36.500 --> 05:43.420
This virtual IP address also is going to need a virtual MAC address.

92
05:43.420 --> 05:48.640
This is the glue that holds all of this together because all of this

93
05:48.640 --> 05:50.860
communication here to get

94
05:50.860 --> 05:55.020
up to the SVI, even a shared virtual IP address.

95
05:55.020 --> 05:59.340
This is all layer 2, which means that we're dealing with MAC addresses.

96
05:59.340 --> 06:04.370
I never know from this host on the left to this host on the right, I never
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

96
05:59.340 --> 06:04.370
I never know from this host on the left to this host on the right, I never

97
06:04.370 --> 06:05.580
learn the other

98
06:05.580 --> 06:10.740
hosts MAC address because it's on a completely different network segment.

99
06:10.740 --> 06:15.780
Instead I'm going to learn the MAC address of my default gateway, which used to

100
06:15.780 --> 06:16.260
be right

101
06:16.260 --> 06:19.860
here physically, but now it's up here virtually.

102
06:19.860 --> 06:25.200
If I'm going to communicate, as we mentioned, using by arping for .254, then I

103
06:25.200 --> 06:25.700
'm going

104
06:25.700 --> 06:31.420
to need to learn the MAC address, which is going to have to also be shared.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

104
06:25.700 --> 06:31.420
to need to learn the MAC address, which is going to have to also be shared.

105
06:31.420 --> 06:36.910
It's going to be shared by our two nodes and it's going to map to the virtual

106
06:36.910 --> 06:38.060
IP address.

107
06:38.060 --> 06:42.420
Everything is in steady state operation and one of these nodes is in charge.

108
06:42.420 --> 06:47.340
Let's say that our active node is here on the left.

109
06:47.340 --> 06:51.100
Then all of a sudden there's some kind of failure.

110
06:51.100 --> 06:56.600
This node goes down and we're going to find that the failover node takes over

111
06:56.600 --> 06:58.300
for .254.

112
06:58.300 --> 07:02.530
Even though we don't have the lines drawn on here, we do have a network path to
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

112
06:58.300 --> 07:02.530
Even though we don't have the lines drawn on here, we do have a network path to

113
07:02.530 --> 07:02.900
this other

114
07:02.900 --> 07:05.380
device.

115
07:05.380 --> 07:09.590
Because we take over for the virtual IP address and the virtual MAC address,

116
07:09.590 --> 07:10.580
now this device

117
07:10.580 --> 07:16.580
takes over and it can run as .254, this host down here never knew there was a

118
07:16.580 --> 07:17.220
failure

119
07:17.220 --> 07:22.260
because we fail over so quickly to the new node.

120
07:22.260 --> 07:27.760
Then we have to have a conversation about what happens now when the active node

121
07:27.760 --> 07:28.260
comes

122
07:28.260 --> 07:30.340
back online.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

122
07:28.260 --> 07:30.340
back online.

123
07:30.340 --> 07:32.500
This is an interesting concept as well.

124
07:32.500 --> 07:40.180
We have the concept of preemption to help us address this, which is basically

125
07:40.180 --> 07:43.340
our failback.

126
07:43.340 --> 07:47.180
The question of whether we want to perform a failback.

127
07:47.180 --> 07:52.510
If the active node is on the left and it experiences a failure and now the

128
07:52.510 --> 07:53.860
active node is on the

129
07:53.860 --> 07:59.130
right, but then this node comes back online, do we want to fail back or are we

130
07:59.130 --> 07:59.840
content

131
07:59.840 --> 08:04.420
just to let this node run as the active until it experiences a failure.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

131
07:59.840 --> 08:04.420
just to let this node run as the active until it experiences a failure.

132
08:04.420 --> 08:09.920
We'll see as we look at HSRP and VRRP and these different FHRP's, the different

133
08:09.920 --> 08:10.780
protocols

134
08:10.780 --> 08:15.580
that we can deploy, that preemption is going to be a topic of conversation.

135
08:15.580 --> 08:20.510
It actually is going to be very important when we get into deeper FHRP designs,

136
08:20.510 --> 08:20.780
but we

137
08:20.780 --> 08:24.420
need to be aware that this concept does exist.

138
08:24.420 --> 08:31.440
If I have preemption enabled on the active, then once it goes down, we have
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

138
08:24.420 --> 08:31.440
If I have preemption enabled on the active, then once it goes down, we have

139
08:31.440 --> 08:32.660
active over

140
08:32.660 --> 08:38.340
here now running and we come back online, well at that point we're going to pre

141
08:38.340 --> 08:38.940
empt the

142
08:38.940 --> 08:42.860
other device and we're going to come back to being the active node.

143
08:42.860 --> 08:45.980
Does this matter from a user perspective?

144
08:45.980 --> 08:48.500
Does it ever know that all of this is happening?

145
08:48.500 --> 08:54.150
No, the endpoints that exist on our network, they are completely oblivious to

146
08:54.150 --> 08:55.260
all of this.

147
08:55.260 --> 09:01.940
This is known as a network level of redundancy, but network is handling this.
WEBVTT
X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000

147
08:55.260 --> 09:01.940
This is known as a network level of redundancy, but network is handling this.

148
09:01.940 --> 09:05.570
We don't have to worry about configuring our endpoints to deal with any of this

149
09:05.570 --> 09:05.820
.

150
09:05.820 --> 09:15.820
[BLANK_AUDIO]