1 00:00:01,684 --> 00:00:04,500 So we got our cluster and we are able 2 00:00:04,500 --> 00:00:06,660 to use cube control with it. 3 00:00:06,660 --> 00:00:08,830 One important thing is missing though. 4 00:00:08,830 --> 00:00:12,430 We got the general cluster network and so on, 5 00:00:12,430 --> 00:00:16,120 but the concrete worker nodes are missing. 6 00:00:16,120 --> 00:00:19,530 And to add them, go to the compute section here 7 00:00:19,530 --> 00:00:23,760 on your cluster and add a node group, click on, 8 00:00:23,760 --> 00:00:27,940 Add node croup and give this node group a name, 9 00:00:27,940 --> 00:00:30,680 for example, demo depth nodes. 10 00:00:30,680 --> 00:00:32,493 But this name is totally up to you. 11 00:00:33,740 --> 00:00:36,950 Next we need to add an IAM role. 12 00:00:36,950 --> 00:00:39,960 And we did this earlier already, for the cluster 13 00:00:39,960 --> 00:00:43,720 to define what the cluster may do on our behalf. 14 00:00:43,720 --> 00:00:46,410 Now we need to do something similar again, 15 00:00:46,410 --> 00:00:49,380 this time however, for the nodes, 16 00:00:49,380 --> 00:00:53,190 because you have to keep in mind that the cluster needed 17 00:00:53,190 --> 00:00:57,050 certain permissions to do some stuff behind the scenes, 18 00:00:57,050 --> 00:00:59,180 the nodes, the worker nodes, 19 00:00:59,180 --> 00:01:02,380 which are in the end EC2 instances 20 00:01:02,380 --> 00:01:04,450 also need certain permissions. 21 00:01:04,450 --> 00:01:07,850 For examples, so that they can write log files 22 00:01:07,850 --> 00:01:10,650 or connect to certain over services, 23 00:01:10,650 --> 00:01:12,970 which we might leverage in our code. 24 00:01:12,970 --> 00:01:16,920 And therefore we need to attach an IAM role 25 00:01:17,847 --> 00:01:21,280 to this node group and therefore to the EC2 instances, 26 00:01:21,280 --> 00:01:22,980 the remote machines, 27 00:01:22,980 --> 00:01:26,140 which will be launched in this node group. 28 00:01:26,140 --> 00:01:30,460 Now the launching will happen automatically managed by AWS, 29 00:01:30,460 --> 00:01:32,480 but we need to set up the permissions 30 00:01:32,480 --> 00:01:35,900 these instances will have, and for this again, 31 00:01:35,900 --> 00:01:40,900 open the IAM console in a new tab and create another role 32 00:01:42,700 --> 00:01:47,370 here, select EC2 on common use cases 33 00:01:47,370 --> 00:01:50,320 and click Next permissions. 34 00:01:50,320 --> 00:01:53,453 And now here are a couple of permissions you need to add, 35 00:01:54,290 --> 00:01:59,290 you should search for EKS here and then EKS worker 36 00:01:59,430 --> 00:02:02,980 and add this EKS worker node policy, 37 00:02:02,980 --> 00:02:05,990 which holds a couple of permissions worker nodes, 38 00:02:05,990 --> 00:02:09,363 which is in the end, what we're going to start here, neat. 39 00:02:11,300 --> 00:02:16,300 Also this CNI policy by searching for a CNI. 40 00:02:17,670 --> 00:02:22,670 And add the ECQ container registry read only policy. 41 00:02:25,610 --> 00:02:29,020 Now these are free policies with a bunch of permissions, 42 00:02:29,020 --> 00:02:32,430 which allow these nodes, these EC2 instances, 43 00:02:32,430 --> 00:02:35,990 which are part of our cluster to do various things, 44 00:02:35,990 --> 00:02:39,430 which they need to do in order to pull images 45 00:02:39,430 --> 00:02:41,083 and to run successfully. 46 00:02:42,010 --> 00:02:44,870 Once you added these free policies, 47 00:02:44,870 --> 00:02:48,660 click on Next tax and Next review, 48 00:02:48,660 --> 00:02:51,170 make sure the free policies are there 49 00:02:51,170 --> 00:02:55,553 and give this role a name like EKS node group, 50 00:02:57,040 --> 00:02:58,670 something like this. 51 00:02:58,670 --> 00:03:01,670 Once you debt click Create role. 52 00:03:01,670 --> 00:03:02,710 And once that is done, 53 00:03:02,710 --> 00:03:06,870 you can close that IAM console and then click 54 00:03:06,870 --> 00:03:09,100 the Refresh button here in your cluster 55 00:03:09,100 --> 00:03:11,440 node group configuration screen, 56 00:03:11,440 --> 00:03:14,533 and then pick this newly created node group here. 57 00:03:16,700 --> 00:03:20,380 With that done you can leave all the other defaults 58 00:03:20,380 --> 00:03:21,743 and click Next. 59 00:03:22,940 --> 00:03:24,010 Now on this page, 60 00:03:24,010 --> 00:03:28,100 you control which kind of EC2 instances will be launched 61 00:03:28,100 --> 00:03:32,630 and managed by AWS on your behalf. 62 00:03:32,630 --> 00:03:35,650 You can keep the default image here 63 00:03:35,650 --> 00:03:37,300 for the operating system, 64 00:03:37,300 --> 00:03:41,460 but for the instance type I'm going to use t3 micro, 65 00:03:41,460 --> 00:03:44,720 which is simply a cheaper instance, 66 00:03:44,720 --> 00:03:47,969 actually make sure you use t3 small 67 00:03:47,969 --> 00:03:51,470 as a minimum here, not micro. 68 00:03:51,470 --> 00:03:54,300 Scheduling ports on micro can fail 69 00:03:54,300 --> 00:03:56,120 and your application can end up 70 00:03:56,120 --> 00:03:58,310 in a pending state, therefore. 71 00:03:58,310 --> 00:04:00,360 So that's just an important node. 72 00:04:00,360 --> 00:04:04,640 Make sure you use the small instance as a minimum value 73 00:04:04,640 --> 00:04:08,290 here not micro, I picked micro here. 74 00:04:08,290 --> 00:04:11,250 I ran into errors and changed it behind the scenes later. 75 00:04:11,250 --> 00:04:15,410 So don't even go down that route, start with small. 76 00:04:15,410 --> 00:04:18,920 The larger the instance here, the more powerful it is, 77 00:04:18,920 --> 00:04:22,750 you can see the number of vertual CPU and memory here, 78 00:04:22,750 --> 00:04:24,940 but of course the more it will cost. 79 00:04:24,940 --> 00:04:28,440 So I will go for the cheapest one here, keep that default, 80 00:04:28,440 --> 00:04:33,440 disc size, and then you can control your scaling policy. 81 00:04:33,950 --> 00:04:37,110 Now, scaling is not something we will look into here, 82 00:04:37,110 --> 00:04:40,160 but I'm happy with starting two nodes 83 00:04:40,160 --> 00:04:41,800 and that's by the way important. 84 00:04:41,800 --> 00:04:44,130 Now we're talking about nodes here. 85 00:04:44,130 --> 00:04:48,140 Not ports, you can't have more than two ports, 86 00:04:48,140 --> 00:04:50,870 but the nodes are just the physical machines 87 00:04:50,870 --> 00:04:52,820 on which your ports run. 88 00:04:52,820 --> 00:04:56,570 With mini cube, we had just one dummy node 89 00:04:56,570 --> 00:04:59,000 in that virtual machine. 90 00:04:59,000 --> 00:05:03,670 Now we have two actual nodes, two different computers, 91 00:05:03,670 --> 00:05:07,530 different machines, so to say across which our ports 92 00:05:07,530 --> 00:05:09,399 and containers will be distributed 93 00:05:09,399 --> 00:05:12,190 automatically by Kubernetes. 94 00:05:12,190 --> 00:05:15,050 that's the part Kubernetes will take care of. 95 00:05:15,050 --> 00:05:18,320 It will start the ports, manage the ports 96 00:05:18,320 --> 00:05:20,893 and distribute them across these nodes. 97 00:05:21,750 --> 00:05:23,310 And the more nodes you have, 98 00:05:23,310 --> 00:05:26,830 the more space you have for the ports to be distributed 99 00:05:26,830 --> 00:05:30,170 so that if the ports on one node have a lot of work 100 00:05:30,170 --> 00:05:33,940 at the moment, new ports could be started on another node. 101 00:05:33,940 --> 00:05:37,020 But of course the more it will cost as well. 102 00:05:37,020 --> 00:05:40,070 So I will go with two nodes here for this demo 103 00:05:40,070 --> 00:05:41,560 and click Next. 104 00:05:41,560 --> 00:05:45,333 Now the settings here should all be fine. 105 00:05:46,320 --> 00:05:49,410 You can disable the remote access to nodes here, 106 00:05:49,410 --> 00:05:52,920 if you want to and I will, then you will not be able 107 00:05:52,920 --> 00:05:56,710 to connect to these nodes to the EC2 instances directly 108 00:05:56,710 --> 00:05:59,090 with SSH, but that's fine for me 109 00:05:59,090 --> 00:06:01,160 because I don't wanna connect anyways. 110 00:06:01,160 --> 00:06:04,800 I wanna let EKS manage my nodes for me, 111 00:06:04,800 --> 00:06:09,130 and therefore I will disable this enter, then click Next 112 00:06:09,130 --> 00:06:12,770 again, we can have another look at all our settings here 113 00:06:12,770 --> 00:06:14,393 and then click Create. 114 00:06:15,280 --> 00:06:19,300 And this will now spin up a couple of EC2 instances 115 00:06:19,300 --> 00:06:22,420 and add them to this cluster. 116 00:06:22,420 --> 00:06:25,150 And of course it will not just launch these instances, 117 00:06:25,150 --> 00:06:30,150 instead, EKS will also install all the Kubernetes software, 118 00:06:30,750 --> 00:06:34,530 which is required like to cube LEDs and cube proxy 119 00:06:34,530 --> 00:06:36,490 on these nodes. 120 00:06:36,490 --> 00:06:39,600 And it will add them all to this cluster network 121 00:06:39,600 --> 00:06:43,710 and do all the heavy lifting of wiring these things up. 122 00:06:43,710 --> 00:06:45,490 And that's the part you would have 123 00:06:45,490 --> 00:06:48,090 to do on your own manually, otherwise, 124 00:06:48,090 --> 00:06:50,990 here, we just specify how many nodes we want 125 00:06:50,990 --> 00:06:53,620 and they are launched and configured for us. 126 00:06:53,620 --> 00:06:56,150 And we don't need to worry about that. 127 00:06:56,150 --> 00:06:58,250 Now this will take a short while 128 00:06:58,250 --> 00:07:00,310 until this is all up and running, 129 00:07:00,310 --> 00:07:04,160 and therefore let's wait until this finished. 130 00:07:04,160 --> 00:07:07,090 And then let's see if we can successfully deploy 131 00:07:07,090 --> 00:07:10,560 our application here with these two Yamel files 132 00:07:10,560 --> 00:07:11,773 onto this cluster. 133 00:07:13,080 --> 00:07:16,168 Now, after a short while, couple of minutes, 134 00:07:16,168 --> 00:07:17,820 this should be active. 135 00:07:17,820 --> 00:07:22,000 Your node group should be active and you can, by the way, 136 00:07:22,000 --> 00:07:26,800 go to the EC2 service page in a new tab, for example, 137 00:07:26,800 --> 00:07:30,210 and you will see you're running EC2 instances here. 138 00:07:30,210 --> 00:07:33,830 These are the instances which were created for us 139 00:07:33,830 --> 00:07:36,640 automatically by EKS. 140 00:07:36,640 --> 00:07:41,020 And that's what I meant when I said that EKS will work 141 00:07:41,020 --> 00:07:43,488 together with other services, 142 00:07:43,488 --> 00:07:48,140 AWS offers in order to bring your EKS cluster to life. 143 00:07:48,140 --> 00:07:51,670 Here, for example, it launched EC2 instances, 144 00:07:51,670 --> 00:07:54,393 through the EC2 service for us. 145 00:07:55,810 --> 00:07:59,000 Whilst you're on that EC2 dashboard, by the way, 146 00:07:59,000 --> 00:08:01,080 also have a look at load balancers, 147 00:08:01,080 --> 00:08:05,110 and please node that you will not see any load balancers 148 00:08:05,110 --> 00:08:07,430 here at the moment. 149 00:08:07,430 --> 00:08:09,363 This will become important later. 150 00:08:10,760 --> 00:08:14,020 So let's close this and go back to our cluster. 151 00:08:14,020 --> 00:08:16,350 Now the cluster is up and running. 152 00:08:16,350 --> 00:08:18,910 It has a network setup for it, 153 00:08:18,910 --> 00:08:22,600 and it has nodes running in that network with all the 154 00:08:22,600 --> 00:08:25,420 Kurbenetes tools installed on them. 155 00:08:25,420 --> 00:08:28,190 And that is all we need to do. 156 00:08:28,190 --> 00:08:33,190 with this our cluster is set up and now it's like mini cube, 157 00:08:33,590 --> 00:08:36,690 just that it's not running in a virtual machine 158 00:08:36,690 --> 00:08:39,070 on our local system, 159 00:08:39,070 --> 00:08:42,309 but it now is a cluster to which we send commands 160 00:08:42,309 --> 00:08:44,360 with the cube control command 161 00:08:44,360 --> 00:08:46,593 just as we did it with mini cube.