1 00:00:02,200 --> 00:00:03,300 Now let's talk 2 00:00:03,300 --> 00:00:06,610 about a problem we might be facing here. 3 00:00:06,610 --> 00:00:10,030 This public IP address changes every time 4 00:00:10,030 --> 00:00:13,090 we deploy an updated container. 5 00:00:13,090 --> 00:00:16,830 And that's, of course, not really ideal. 6 00:00:16,830 --> 00:00:19,900 Now, that's why I did create that load balancer. 7 00:00:19,900 --> 00:00:22,080 If you go to Services and you search 8 00:00:22,080 --> 00:00:24,913 for application load here, 9 00:00:26,000 --> 00:00:28,600 you should find EC2 as a suggestion 10 00:00:28,600 --> 00:00:32,350 because that's where load balancers are managed. 11 00:00:32,350 --> 00:00:34,730 We don't use EC2 instances here 12 00:00:34,730 --> 00:00:37,590 but the load balancer feature belongs to that 13 00:00:37,590 --> 00:00:40,350 and Fargate also utilizes it. 14 00:00:40,350 --> 00:00:42,330 So here under Load Balancers, 15 00:00:42,330 --> 00:00:45,394 you should find the load balancer we created. 16 00:00:45,394 --> 00:00:48,510 And here, we find the DNS name, 17 00:00:48,510 --> 00:00:51,010 which looks very much like a domain. 18 00:00:51,010 --> 00:00:52,530 We can copy that, 19 00:00:52,530 --> 00:00:53,970 enter it in the browser 20 00:00:53,970 --> 00:00:57,210 and send a request to /goals. 21 00:00:57,210 --> 00:01:00,770 However, you will notice that this won't work here 22 00:01:00,770 --> 00:01:03,160 for a couple of reasons. 23 00:01:03,160 --> 00:01:05,190 For one, I made a tiny mistake 24 00:01:05,190 --> 00:01:08,680 in the target groups of that load balancer. 25 00:01:08,680 --> 00:01:10,580 If we go back to ECS, 26 00:01:10,580 --> 00:01:13,970 you will notice that on your services, 27 00:01:13,970 --> 00:01:17,010 the tasks connected to the service 28 00:01:17,010 --> 00:01:20,880 are stopped and restarted all the time. 29 00:01:20,880 --> 00:01:22,906 And that stopping and restarting 30 00:01:22,906 --> 00:01:26,010 is done by our load balancer 31 00:01:26,010 --> 00:01:28,880 because that load balancer is checking the health 32 00:01:28,880 --> 00:01:32,800 of the services it's redirecting requests to 33 00:01:32,800 --> 00:01:34,470 and it shuts down a service 34 00:01:34,470 --> 00:01:39,020 and restarts it if finds that this service is unhealthy. 35 00:01:39,020 --> 00:01:41,840 For example, if it's not responding anymore. 36 00:01:41,840 --> 00:01:44,410 That generally is a good behavior 37 00:01:44,410 --> 00:01:48,100 because it ensures that if our application should crash 38 00:01:48,100 --> 00:01:50,980 for whatever reason, it's restarted. 39 00:01:50,980 --> 00:01:53,720 However, unfortunately, I made a tiny mistake 40 00:01:53,720 --> 00:01:56,580 when we set up this load balancer. 41 00:01:56,580 --> 00:02:01,080 On the target groups area of that EC2 service page, 42 00:02:01,080 --> 00:02:04,830 you should find the target group we created earlier. 43 00:02:04,830 --> 00:02:07,890 There under Health check settings, click on Edit. 44 00:02:07,890 --> 00:02:10,729 And here, by default, the load balancer 45 00:02:10,729 --> 00:02:15,730 is sending a request to our deployed service slash nothing. 46 00:02:17,590 --> 00:02:22,590 Now, in our case, we are handling requests to /goals 47 00:02:22,610 --> 00:02:25,320 but not to slash nothing. 48 00:02:25,320 --> 00:02:29,010 Therefore, requests sent to slash nothing 49 00:02:29,010 --> 00:02:32,430 will respond with a 404 error, 50 00:02:32,430 --> 00:02:34,573 which is treated as unhealthy. 51 00:02:35,610 --> 00:02:40,000 Therefore, we should change this health check path to /goals 52 00:02:40,000 --> 00:02:43,220 so that the request is sent to an endpoint 53 00:02:43,220 --> 00:02:45,740 of our service, which will respond 54 00:02:45,740 --> 00:02:47,560 with a success code. 55 00:02:47,560 --> 00:02:49,530 Now, that's one thing we've gotta change 56 00:02:49,530 --> 00:02:52,653 to ensure that our load balancer will work correctly. 57 00:02:53,510 --> 00:02:57,030 Another thing which I accidentally messed up a bit 58 00:02:57,030 --> 00:02:59,400 is on the load balancer itself 59 00:02:59,400 --> 00:03:02,430 in the basic configuration here, 60 00:03:02,430 --> 00:03:04,530 the security groups I added, 61 00:03:04,530 --> 00:03:07,490 I added this default security group. 62 00:03:07,490 --> 00:03:11,990 Now, actually I wanna add this goals security groups here, 63 00:03:11,990 --> 00:03:13,690 not the default one. 64 00:03:13,690 --> 00:03:16,710 So here, make sure you add your security group 65 00:03:16,710 --> 00:03:19,290 besides that default security group 66 00:03:19,290 --> 00:03:21,660 so that the security group we're also using 67 00:03:21,660 --> 00:03:25,000 for our ECS services is also attached here 68 00:03:25,000 --> 00:03:26,383 and then click Save. 69 00:03:27,960 --> 00:03:30,510 Now, all these changes will take a couple of seconds 70 00:03:30,510 --> 00:03:32,480 or minutes to have an effect 71 00:03:32,480 --> 00:03:36,300 but thereafter, our services should not be stopped 72 00:03:36,300 --> 00:03:38,300 and restarted all the time anymore 73 00:03:38,300 --> 00:03:40,750 or our tasks here, I should say. 74 00:03:40,750 --> 00:03:42,230 And we should be able 75 00:03:42,230 --> 00:03:45,340 to also use this URL here, 76 00:03:45,340 --> 00:03:47,960 which we find on the load balancer page 77 00:03:47,960 --> 00:03:50,920 to send a request to our deployed services 78 00:03:50,920 --> 00:03:52,850 and to our running tasks. 79 00:03:52,850 --> 00:03:56,070 And that means that we now can send a request 80 00:03:56,070 --> 00:03:59,460 to the same URL, to the same domain 81 00:03:59,460 --> 00:04:03,320 without caring about changing IP addresses. 82 00:04:03,320 --> 00:04:06,380 Of course, you can also map your own custom domain 83 00:04:06,380 --> 00:04:08,100 to this one if you want to 84 00:04:08,100 --> 00:04:11,510 and the AWS documentation is the place to go 85 00:04:11,510 --> 00:04:13,990 if you wanna learn more about that. 86 00:04:13,990 --> 00:04:16,459 So if we now wait for all this deployment 87 00:04:16,459 --> 00:04:17,866 to finish up here, 88 00:04:17,866 --> 00:04:22,866 and we wait for this service to be restarted one more time, 89 00:04:23,247 --> 00:04:25,990 and therefore, after a couple of minutes, 90 00:04:25,990 --> 00:04:27,920 we will be able to send the request 91 00:04:27,920 --> 00:04:30,063 to this domain /goals. 92 00:04:31,042 --> 00:04:34,220 And of course, we can also use this in Postman now 93 00:04:34,220 --> 00:04:36,040 instead of the IP address 94 00:04:36,040 --> 00:04:37,699 and we can add a new goal. 95 00:04:37,699 --> 00:04:41,136 We can, of course, then also get our goals. 96 00:04:41,136 --> 00:04:43,530 Oops, I had a typo there. 97 00:04:43,530 --> 00:04:45,170 We can get our goals 98 00:04:45,170 --> 00:04:47,210 and of course, if we want to, 99 00:04:47,210 --> 00:04:50,010 we can also delete a goal there. 100 00:04:50,010 --> 00:04:53,050 Just make sure you grab the correct goal ID, 101 00:04:53,050 --> 00:04:54,370 plug it into the URL 102 00:04:55,360 --> 00:04:57,490 and here we go. 103 00:04:57,490 --> 00:04:59,460 So this is now all working 104 00:04:59,460 --> 00:05:02,800 and that's how we could deploy multiple containers 105 00:05:02,800 --> 00:05:04,630 with AWS ECS 106 00:05:04,630 --> 00:05:07,740 and how we can also use such a load balancer 107 00:05:07,740 --> 00:05:11,060 to then ensure that we have a unchanging domain, 108 00:05:11,060 --> 00:05:13,793 which points at these deployed containers.