WEBVTT 1 00:00.210 --> 00:01.470 So let's take a closer look 2 00:01.470 --> 00:03.450 at what exactly is a transport node. 3 00:03.450 --> 00:04.770 And again, the high level overview 4 00:04.770 --> 00:07.980 is that a transport node is an ESXi host 5 00:07.980 --> 00:11.160 that we know and love where we've added to it 6 00:11.160 --> 00:13.710 the ability to work with NSX. 7 00:13.710 --> 00:14.670 So the way that would look 8 00:14.670 --> 00:17.040 is you and I would go to our NSX Manager. 9 00:17.040 --> 00:20.430 The NSX Manager is working in conjunction with vCenter, 10 00:20.430 --> 00:23.370 so it knows exactly about which hosts are available. 11 00:23.370 --> 00:25.710 And then using NSX Manager, we reach out to a host, 12 00:25.710 --> 00:28.410 let's say ESXi-A, and we specify 13 00:28.410 --> 00:31.140 that we wanted to participate as part of NSX. WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 13 00:28.410 --> 00:31.140 that we wanted to participate as part of NSX. 14 00:31.140 --> 00:33.810 And once we do, the NSX software is installed 15 00:33.810 --> 00:36.630 and enabled on ESXi-A and bada bing, bada boom, 16 00:36.630 --> 00:38.700 it becomes a transport node. 17 00:38.700 --> 00:40.620 And part of that, the ESXi host now 18 00:40.620 --> 00:42.600 is gonna get a new VMkernel adapter 19 00:42.600 --> 00:44.730 and is gonna use that VMkernel adapter 20 00:44.730 --> 00:47.430 as one end of its tunnel called a tunnel endpoint, 21 00:47.430 --> 00:49.290 so that if it needs to, it can encapsulate 22 00:49.290 --> 00:53.160 and forward traffic over that tunnel to another entity. 23 00:53.160 --> 00:53.993 And up to this point, 24 00:53.993 --> 00:55.020 we haven't talked about other nodes 25 00:55.020 --> 00:55.910 other than a transport node. 26 00:55.910 --> 00:57.660 So let's imagine it sending it over 27 00:57.660 --> 01:00.123 to another ESXi host like ESXi-B, WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 27 00:57.660 --> 01:00.123 to another ESXi host like ESXi-B, 28 01:01.063 --> 01:02.052 who could then deencapsulate it 29 01:02.052 --> 01:03.450 and then forward it onto to its final destination. 30 01:03.450 --> 01:04.560 And for every good tunnel, 31 01:04.560 --> 01:07.020 there's two tunnel endpoints, one at each side. 32 01:07.020 --> 01:08.730 So if we had two clients, for example, 33 01:08.730 --> 01:10.980 right here on segment 20, 34 01:10.980 --> 01:13.920 and this one was running on ESXi-A, 35 01:13.920 --> 01:15.690 and this one was running on ESXi-B, 36 01:15.690 --> 01:17.490 and they're both connected to that same segment 37 01:17.490 --> 01:18.810 as they communicated back and forth, 38 01:18.810 --> 01:20.430 the traffic would be encapsulated. 39 01:20.430 --> 01:21.980 If this is VMA, and this is VMB 40 01:22.950 --> 01:24.510 and VMA is communicating with VMB, 41 01:24.510 --> 01:27.060 the traffic would be initiated by VMA, 42 01:27.060 --> 01:28.620 ESXi-A would encapsulate it, 43 01:28.620 --> 01:30.690 send it through the tunnel as a routed packet WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 43 01:28.620 --> 01:30.690 send it through the tunnel as a routed packet 44 01:30.690 --> 01:32.910 over to ESXi-B, who would deencapsulate it, 45 01:32.910 --> 01:34.710 and then forwarded over to VMB, 46 01:34.710 --> 01:36.150 which is one of the great reasons why 47 01:36.150 --> 01:37.770 it's called a transport node. 48 01:37.770 --> 01:40.050 It's able to do transport over a tunnel, 49 01:40.050 --> 01:41.640 specifically a GENEVE tunnel 50 01:41.640 --> 01:43.680 from itself to the other end point. 51 01:43.680 --> 01:44.940 So let me clean this up a little bit 52 01:44.940 --> 01:47.100 and let's chat about a few additional terms. 53 01:47.100 --> 01:49.680 One of those terms is called a transport zone. 54 01:49.680 --> 01:52.320 And just for fun, let me go ahead and draw a couple zones. 55 01:52.320 --> 01:53.640 I'm gonna put a zone here, 56 01:53.640 --> 01:54.990 and then I'll put another zone here 57 01:54.990 --> 01:55.823 and let me label those. 58 01:55.823 --> 01:57.900 Let's call this zone right here the production zone, 59 01:57.900 --> 01:59.700 I'll call it prod zone for short. 60 01:59.700 --> 02:01.950 And over here, I'll call this the dev zone, WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 60 01:59.700 --> 02:01.950 And over here, I'll call this the dev zone, 61 02:01.950 --> 02:03.210 short for development. 62 02:03.210 --> 02:05.370 Let's imagine we have six ESXi hosts 63 02:05.370 --> 02:08.040 that we've enabled for NSX. 64 02:08.040 --> 02:09.600 So if they're enabled for NSX, 65 02:09.600 --> 02:12.990 those ESXi hosts are referred to now as transport nodes 66 02:12.990 --> 02:14.850 for the purposes of NSX. 67 02:14.850 --> 02:16.560 So I'll put four of 'em right here. 68 02:16.560 --> 02:18.690 I'll label them one through four. 69 02:18.690 --> 02:21.660 Those are ESXi hosts 1, 2, 3, and 4. 70 02:21.660 --> 02:22.590 And over here, 71 02:22.590 --> 02:24.870 I'll specify that these other two hosts 72 02:24.870 --> 02:26.550 are gonna be in the dev zone only. 73 02:26.550 --> 02:28.680 So those are host 5 and 6. 74 02:28.680 --> 02:31.500 So what exactly is a transport zone good for? WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 74 02:28.680 --> 02:31.500 So what exactly is a transport zone good for? 75 02:31.500 --> 02:35.130 Transport zones control the scope of a new segment. 76 02:35.130 --> 02:36.720 So let's create a new segment, 77 02:36.720 --> 02:38.400 and we'll create this layer two segment, 78 02:38.400 --> 02:40.470 we'll call it segment 50. 79 02:40.470 --> 02:45.030 So this NSX segment is a logical layer two broadcast domain. 80 02:45.030 --> 02:48.300 Now, it could exist across all of our ESXi hosts. 81 02:48.300 --> 02:50.910 However, if we've carved up these four hosts, 82 02:50.910 --> 02:52.920 so they're only participating in the production zone, 83 02:52.920 --> 02:55.320 and these two hosts are only participating in the dev zone, 84 02:55.320 --> 02:56.340 when we create a new segment, 85 02:56.340 --> 02:59.010 we can associate that segment with the production zone. 86 02:59.010 --> 03:01.830 And that way the segment effectively would be reachable WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 86 02:59.010 --> 03:01.830 And that way the segment effectively would be reachable 87 03:01.830 --> 03:05.760 and accessible as a port group for these four ESXi hosts. 88 03:05.760 --> 03:08.580 But if segment 50 is not associated with the dev zone, 89 03:08.580 --> 03:12.000 these two ESXi hosts are not gonna have that port group. 90 03:12.000 --> 03:13.290 Or in the terms of NSX, 91 03:13.290 --> 03:15.360 they're not gonna have that logical segment 92 03:15.360 --> 03:18.000 as part of their virtual distributed switch. 93 03:18.000 --> 03:20.220 And just to be fair, let's create another segment. 94 03:20.220 --> 03:21.540 We'll go ahead and draw it like this, 95 03:21.540 --> 03:23.340 and we'll call it segment 60. 96 03:23.340 --> 03:25.290 Let's imagine we only have this segment available 97 03:25.290 --> 03:26.123 for these two hosts. 98 03:26.123 --> 03:27.480 So what we could do is we could go ahead 99 03:27.480 --> 03:29.760 and associate the segment with the dev zone. 100 03:29.760 --> 03:31.680 And because these two hosts are participating WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 100 03:29.760 --> 03:31.680 And because these two hosts are participating 101 03:31.680 --> 03:34.200 in the dev zone, only these two hosts 102 03:34.200 --> 03:35.820 would have that segment available to 'em. 103 03:35.820 --> 03:38.340 Again, it would appear on the virtual distributed switch. 104 03:38.340 --> 03:41.100 It simply appears like a logical port group 105 03:41.100 --> 03:42.180 that is provided by NSX. 106 03:42.180 --> 03:43.860 So it's referred to as a segment. 107 03:43.860 --> 03:46.650 So when we think about the concept of a transport zone, 108 03:46.650 --> 03:48.780 think about the concept of scope, 109 03:48.780 --> 03:50.580 meaning how many ESXi hosts 110 03:50.580 --> 03:53.850 and how many devices are gonna be supporting that segment. 111 03:53.850 --> 03:55.380 So if a segment, again, is associated 112 03:55.380 --> 03:56.490 with the production zone, 113 03:56.490 --> 03:58.770 it's only being available to these four hosts. 114 03:58.770 --> 04:00.900 If we have a segment associated with the dev zone, WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 114 03:58.770 --> 04:00.900 If we have a segment associated with the dev zone, 115 04:00.900 --> 04:02.250 it would then only be available 116 04:02.250 --> 04:03.840 to these two transport nodes. 117 04:03.840 --> 04:07.200 Also, regarding ESXi hosts that have been enabled for NSX, 118 04:07.200 --> 04:08.550 not only are they transport node, 119 04:08.550 --> 04:10.020 but if we're gonna be explicit, 120 04:10.020 --> 04:13.080 we could also call them a host transport node, 121 04:13.080 --> 04:14.610 which is gonna help differentiate them 122 04:14.610 --> 04:16.020 from another type of node 123 04:16.020 --> 04:17.700 we're gonna talk about in the next video. 124 04:17.700 --> 04:19.560 And while we're chatting about host transport nodes, 125 04:19.560 --> 04:21.600 let's also take a look at the uplinks. 126 04:21.600 --> 04:24.420 So let's imagine we have this host, let's call it Host A. 127 04:24.420 --> 04:27.270 And Host A has a few vmnics connected 128 04:27.270 --> 04:28.350 up to the physical world. 129 04:28.350 --> 04:30.900 So here's our physical infrastructure WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 129 04:28.350 --> 04:30.900 So here's our physical infrastructure 130 04:30.900 --> 04:32.730 with some high speed ethernet switching out here. 131 04:32.730 --> 04:34.380 So I'll label this as physical. 132 04:34.380 --> 04:35.213 And for the vmnics, 133 04:35.213 --> 04:38.580 let's call 'em vmnic 0 and 1 and 2 and 3. 134 04:38.580 --> 04:40.260 So when we go to the NSX manager, 135 04:40.260 --> 04:42.510 and we specify that we want Host A to participate 136 04:42.510 --> 04:45.090 and become a host transport node as part of NSX, 137 04:45.090 --> 04:46.890 we are also going to be asked 138 04:46.890 --> 04:48.180 and need to provide information 139 04:48.180 --> 04:51.690 regarding which uplinks we want this host to use 140 04:51.690 --> 04:53.790 as part of being a host transport note. 141 04:53.790 --> 04:57.900 Do we wanna use vmnic 0 and 1, or 2 and 3, or 1 and 3? 142 04:57.900 --> 05:01.590 And do we wanna use teaming in the form of load balancing, WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 142 04:57.900 --> 05:01.590 And do we wanna use teaming in the form of load balancing, 143 05:01.590 --> 05:03.900 or do we wanna use a failover order? 144 05:03.900 --> 05:06.600 And so the details regarding which uplinks we're gonna use 145 05:06.600 --> 05:09.390 are configured as part of an uplink profile. 146 05:09.390 --> 05:12.060 So it'd be a great idea before we start configuring 147 05:12.060 --> 05:15.420 and setting up an ESXi host as a host transport node, 148 05:15.420 --> 05:17.850 we'd wanna plan out and identify beforehand 149 05:17.850 --> 05:21.300 exactly which interfaces we wanna use for the uplinks 150 05:21.300 --> 05:23.610 and our uplink strategy. 151 05:23.610 --> 05:26.280 Again, do we wanna do load balancing with two or more nics 152 05:26.280 --> 05:28.230 or do we wanna have two or more vmnics 153 05:28.230 --> 05:30.000 in a specific failover order? WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 154 05:30.000 --> 05:32.580 And we're gonna specify those as part of an uplink profile 155 05:32.580 --> 05:35.130 when we're configuring our host transport node. 156 05:35.130 --> 05:36.630 And what's really interesting about this 157 05:36.630 --> 05:39.510 is that if we specify one uplink that we're gonna use, 158 05:39.510 --> 05:42.630 it's gonna automatically create a new VMkernel adapter 159 05:42.630 --> 05:44.250 with an IP address that we specify 160 05:44.250 --> 05:45.900 as part of our configuration. 161 05:45.900 --> 05:47.370 And the IP address that we're gonna specify 162 05:47.370 --> 05:50.280 is the tunnel endpoint on this ESXi host. 163 05:50.280 --> 05:52.410 Now, the key is if we did two uplinks 164 05:52.410 --> 05:53.880 as part of our uplink profile, 165 05:53.880 --> 05:55.440 it's not just gonna have one VMK 166 05:55.440 --> 05:57.330 with an IP address for a tunnel endpoint. 167 05:57.330 --> 05:58.950 Instead, it's gonna have two. 168 05:58.950 --> 06:00.120 So you have two uplinks. WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 168 05:58.950 --> 06:00.120 So you have two uplinks. 169 06:00.120 --> 06:03.060 It's gonna equate to two tunnel endpoints that it can use 170 06:03.060 --> 06:05.820 to communicate with other nodes in NSX. 171 06:05.820 --> 06:07.320 And let me show you a quick example of that 172 06:07.320 --> 06:08.670 to reinforce that concept. 173 06:08.670 --> 06:10.290 So this is the NSX Manager, 174 06:10.290 --> 06:12.900 and this is the section for uplink profiles right here. 175 06:12.900 --> 06:16.440 And here's an uplink profile called ESXi-A, et cetera. 176 06:16.440 --> 06:18.690 And it's using exactly one uplink. 177 06:18.690 --> 06:22.800 So as a result, we would expect one tunnel endpoint address 178 06:22.800 --> 06:25.500 on the host transport node that's using this uplink profile. 179 06:25.500 --> 06:28.560 So if we go to our hosts and we go down to ESXi-A 180 06:28.560 --> 06:30.120 and take a look at the details, WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 180 06:28.560 --> 06:30.120 and take a look at the details, 181 06:30.120 --> 06:31.590 and I'm gonna click on Configure 182 06:31.590 --> 06:33.570 and click on Next a couple of times 183 06:33.570 --> 06:34.830 and click on Next. 184 06:34.830 --> 06:37.080 Right here is giving us a visual representation 185 06:37.080 --> 06:38.670 that is using vmnic 2 186 06:38.670 --> 06:40.140 And if we expand this, 187 06:40.140 --> 06:41.940 it's showing us the uplink profile it's using, 188 06:41.940 --> 06:43.680 which is the ESXi-A one. 189 06:43.680 --> 06:46.770 And as a result of that, having one uplink for the NSX, 190 06:46.770 --> 06:48.450 I'll go ahead and click on Cancel here. 191 06:48.450 --> 06:50.340 If we look at the tunnel endpoint address, 192 06:50.340 --> 06:54.510 it has one tunnel endpoint address that's 10.11.11.11 193 06:54.510 --> 06:56.760 which I hard coded for that endpoint. 194 06:56.760 --> 06:59.250 And again, if we had two uplinks or three uplinks, 195 06:59.250 --> 07:01.290 we'd have two or three respectively WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 195 06:59.250 --> 07:01.290 we'd have two or three respectively 196 07:01.290 --> 07:05.010 tunnel endpoint addresses, one for each of our uplinks. 197 07:05.010 --> 07:06.720 And then if we went to the vSphere Client 198 07:06.720 --> 07:08.850 and took a look at the ESXi-A host. 199 07:08.850 --> 07:10.290 So here's the vSphere Client, 200 07:10.290 --> 07:11.670 we're in the host and clusters view, 201 07:11.670 --> 07:13.370 we're looking at ESXi-A, 202 07:13.370 --> 07:14.850 if we go to Configure 203 07:14.850 --> 07:16.650 and then go down to VMkernel adapters. 204 07:16.650 --> 07:19.890 Here is the VMkernel adapter that was created by NSX 205 07:19.890 --> 07:21.810 for the benefit of that tunnel endpoint 206 07:21.810 --> 07:24.510 with that IP address of 10.11.11.11. 207 07:24.510 --> 07:26.160 And again, if we had two uplinks 208 07:26.160 --> 07:28.080 associated with NSX for this host, 209 07:28.080 --> 07:31.020 it would have two specific VMkernel adapters WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 209 07:28.080 --> 07:31.020 it would have two specific VMkernel adapters 210 07:31.020 --> 07:34.500 with two IP addresses, one for each of its tunnels. 211 07:34.500 --> 07:35.333 And again, in this skill, 212 07:35.333 --> 07:37.530 I just wanna reinforce some of the basic concepts 213 07:37.530 --> 07:38.970 behind NSX. 214 07:38.970 --> 07:41.490 So on our Host A, because we specified one uplink, 215 07:41.490 --> 07:44.190 it has one VMkernel adapter specific 216 07:44.190 --> 07:45.630 for that tunnel endpoint. 217 07:45.630 --> 07:47.310 So in the interface, I think it showed us 10, 218 07:47.310 --> 07:51.300 and then it had the IP address of 10.11.11.11 219 07:51.300 --> 07:54.000 which is the tunnel endpoint on this ESXi host 220 07:54.000 --> 07:56.910 that it can use to communicate with other devices 221 07:56.910 --> 07:58.800 who are at the other end of its tunnels. 222 07:58.800 --> 08:01.050 So the other end of the tunnel could be Host B. WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 222 07:58.800 --> 08:01.050 So the other end of the tunnel could be Host B. 223 08:01.050 --> 08:03.450 So if there was Host B, that would be one tunnel 224 08:03.450 --> 08:04.830 and it had yet another host, let's go ahead, 225 08:04.830 --> 08:07.020 but Host C here with one uplink 226 08:07.020 --> 08:08.280 from that tunnel endpoint address, 227 08:08.280 --> 08:09.900 we'd have another tunnel going over to Host C 228 08:09.900 --> 08:11.340 to its tunnel endpoint. 229 08:11.340 --> 08:13.650 And in addition to building these tunnels 230 08:13.650 --> 08:15.870 for the tunnel endpoints to other ESXi hosts, 231 08:15.870 --> 08:17.760 which are also host transport nodes, 232 08:17.760 --> 08:20.670 there's another type of node we can also build a tunnel too, 233 08:20.670 --> 08:22.530 and that's called an edge node. 234 08:22.530 --> 08:23.970 So it could be called an edge node, 235 08:23.970 --> 08:27.120 or it's also referred to as an edge transport node. 236 08:27.120 --> 08:28.230 And that would differentiate it 237 08:28.230 --> 08:31.530 from the host transport node, which is an ESXi host. WEBVTT X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:900000 237 08:28.230 --> 08:31.530 from the host transport node, which is an ESXi host. 238 08:31.530 --> 08:32.430 And the question might be, well, 239 08:32.430 --> 08:35.160 why do we need this thing called an edge node 240 08:35.160 --> 08:36.117 or an edge transport node? 241 08:36.117 --> 08:38.550 And the answer is the edge transport nodes 242 08:38.550 --> 08:40.702 are gonna take the responsibility of implementing 243 08:40.702 --> 08:44.430 our logical layer three gateways and services 244 08:44.430 --> 08:46.480 should we choose to use them for routing.