1 00:00:02,200 --> 00:00:03,850 With that we're almost done 2 00:00:03,850 --> 00:00:07,290 with these very basics about Kubernetes 3 00:00:07,290 --> 00:00:10,260 and resources with Kubernetes. 4 00:00:10,260 --> 00:00:13,220 There's one a little bit more advanced concept, 5 00:00:13,220 --> 00:00:15,120 which I wanna touch on right now. 6 00:00:15,120 --> 00:00:17,750 And that is related to how Kubernetes 7 00:00:17,750 --> 00:00:20,200 checks whether your pods and containers, 8 00:00:20,200 --> 00:00:22,760 and the pods are healthy or not, 9 00:00:22,760 --> 00:00:26,530 because this can matter in certain applications. 10 00:00:26,530 --> 00:00:27,363 Up to this point, 11 00:00:27,363 --> 00:00:30,580 it always works the way it seems to do it by default 12 00:00:30,580 --> 00:00:33,960 but you can of course configure how it should do that. 13 00:00:33,960 --> 00:00:36,520 And for this deployment Yami file, 14 00:00:36,520 --> 00:00:39,430 I will actually go to my container definition 15 00:00:39,430 --> 00:00:43,930 to change how this container is checked by the pod 16 00:00:43,930 --> 00:00:46,690 to see whether it's up and running. 17 00:00:46,690 --> 00:00:48,550 And we can change this 18 00:00:48,550 --> 00:00:53,550 by adding a so called liveness probe key. 19 00:00:54,650 --> 00:00:56,080 And again, you find this 20 00:00:56,080 --> 00:00:58,140 in your official documentation as well. 21 00:00:58,140 --> 00:01:00,010 If you have a look at the deployment, 22 00:01:00,010 --> 00:01:01,770 the deploymentSpec 23 00:01:01,770 --> 00:01:05,000 they are at the template and the podSpec 24 00:01:05,000 --> 00:01:08,000 there if you dive into the container array, 25 00:01:08,000 --> 00:01:11,220 you find that for containers, 26 00:01:11,220 --> 00:01:13,210 you can configure a lot of things. 27 00:01:13,210 --> 00:01:15,250 For example, all the environment variables 28 00:01:15,250 --> 00:01:17,560 which can become important later 29 00:01:17,560 --> 00:01:19,800 also how the image should be pulled 30 00:01:19,800 --> 00:01:22,683 and here you also find the liveness probe key. 31 00:01:24,920 --> 00:01:27,310 So, how can we configure this? 32 00:01:27,310 --> 00:01:30,810 Well, there you can add a httpGet key, 33 00:01:30,810 --> 00:01:32,470 and of course there are more options, 34 00:01:32,470 --> 00:01:34,040 which you can learn about if you dive 35 00:01:34,040 --> 00:01:36,210 into that probe definition 36 00:01:36,210 --> 00:01:38,080 or simply into the official docs. 37 00:01:38,080 --> 00:01:41,560 If you search for kubernetes liveness probe, 38 00:01:41,560 --> 00:01:45,210 here you find more information and documentation, 39 00:01:45,210 --> 00:01:47,560 but here we can simply define 40 00:01:47,560 --> 00:01:50,270 that a Gethttp request should be sent 41 00:01:50,270 --> 00:01:53,210 by the pod to the running application so to say, 42 00:01:53,210 --> 00:01:56,290 or by Kubernetes to the running application. 43 00:01:56,290 --> 00:01:59,630 We wanna send it to a path of just slash, 44 00:01:59,630 --> 00:02:01,930 but of course that could be another nested path 45 00:02:01,930 --> 00:02:03,603 your application might have, 46 00:02:04,600 --> 00:02:07,710 on port 8080 in our container, 47 00:02:07,710 --> 00:02:10,780 since that's the port this container exposes 48 00:02:10,780 --> 00:02:13,240 maybe with some extra Headers, 49 00:02:13,240 --> 00:02:16,260 but we don't need some here so we can omit this. 50 00:02:16,260 --> 00:02:19,220 And then, on the same level as httpGet 51 00:02:19,220 --> 00:02:21,320 not nested inside of it, 52 00:02:21,320 --> 00:02:25,020 you add the periodSeconds value 53 00:02:25,020 --> 00:02:28,360 where you define how often that should be performed. 54 00:02:28,360 --> 00:02:31,120 For example, all three seconds or 10 seconds, 55 00:02:31,120 --> 00:02:32,620 whatever you want. 56 00:02:32,620 --> 00:02:36,690 In addition, you can add to the initial DelaySeconds key, 57 00:02:36,690 --> 00:02:39,820 where you can tell Kubernetes how long it should wait 58 00:02:39,820 --> 00:02:42,520 until it checks the health for the first time, 59 00:02:42,520 --> 00:02:45,040 for example, five seconds. 60 00:02:45,040 --> 00:02:48,300 And now we've got our own liveness probe in action, 61 00:02:48,300 --> 00:02:52,060 where we define how Kubernetes should verify 62 00:02:52,060 --> 00:02:54,923 whether that container is up and running. 63 00:02:56,090 --> 00:02:58,070 And if I now save this 64 00:02:58,070 --> 00:03:03,040 I can of course again apply the deployment.yaml file 65 00:03:03,040 --> 00:03:05,793 and service.yaml file. 66 00:03:07,580 --> 00:03:11,540 And if you then run minikube service backend 67 00:03:11,540 --> 00:03:15,773 we again get our URL to the deployed application. 68 00:03:16,850 --> 00:03:17,900 Here it is. 69 00:03:17,900 --> 00:03:20,150 But if I now visit slash error 70 00:03:21,670 --> 00:03:25,230 it of course crashed, but it's opens immediately 71 00:03:25,230 --> 00:03:28,600 up and running again because that crash was detected 72 00:03:28,600 --> 00:03:31,980 and the container and the pod was restarted 73 00:03:31,980 --> 00:03:34,520 as you can see if you get your pods. 74 00:03:34,520 --> 00:03:37,890 you see it was restarted once here. 75 00:03:37,890 --> 00:03:40,520 Now that is something which worked before as well. 76 00:03:40,520 --> 00:03:41,840 I will say that, 77 00:03:41,840 --> 00:03:45,070 but it still works with our own liveness probe 78 00:03:45,070 --> 00:03:48,520 and defining such a liveness probe can be very useful 79 00:03:48,520 --> 00:03:50,010 if you have an application 80 00:03:50,010 --> 00:03:52,310 which doesn't react to the default, 81 00:03:52,310 --> 00:03:54,620 or if you simply want to check the health 82 00:03:54,620 --> 00:03:57,220 by sending a request to slash something 83 00:03:57,220 --> 00:03:59,300 instead of just slash. 84 00:03:59,300 --> 00:04:02,830 So knowing how that can be defined can be very helpful 85 00:04:02,830 --> 00:04:05,700 to ensure that your applications work 86 00:04:05,700 --> 00:04:09,163 no matter how they are structured or configured.