1 00:00:04,370 --> 00:00:10,700 Well, let's go ahead and use Carl to check the info and health and point toward the respective information, 2 00:00:10,940 --> 00:00:15,860 because this image is based on Alpina Linux, but it doesn't come with Kerl. 3 00:00:16,070 --> 00:00:22,160 So for that reason will have to use the Alpine Linux package manager to install Kerl and to create a 4 00:00:22,320 --> 00:00:26,960 fourth block in order to install these dependencies that are not inside an image. 5 00:00:33,360 --> 00:00:41,250 So this command will install Karl, and now after it has deployed to death, we can use Colonel and 6 00:00:41,250 --> 00:00:47,970 this will be the domain name and we're going to search with GRAP or, for example, the pipeline at. 7 00:00:53,450 --> 00:00:59,240 And the same thing will do in the health and point we're going to search for up. 8 00:01:01,670 --> 00:01:06,620 The next step would be to get the domain name in there multiple ways on how we can do that, wanted 9 00:01:06,620 --> 00:01:12,680 to show you a dynamic way on how we can find out a domain name without defining a variable where you 10 00:01:12,680 --> 00:01:13,880 hardcoded domain name. 11 00:01:14,930 --> 00:01:20,840 The information that we are looking for is already presented after deploying the application, and you 12 00:01:20,840 --> 00:01:27,350 will see here that in this Jason format, the HWC like console gets back to us and says which version 13 00:01:27,350 --> 00:01:29,860 has been deployed and which is the domain name? 14 00:01:30,530 --> 00:01:35,230 Its name will be the canonical name for the domain and you will see its production. 15 00:01:35,240 --> 00:01:42,460 That's something we can later change it to something else and then we will not have to update our pipeline. 16 00:01:43,070 --> 00:01:50,900 That will do next is to pass this June five is tracked this domain name from this property name and 17 00:01:50,900 --> 00:01:53,960 set an environment variable that I'm going to use. 18 00:01:53,960 --> 00:01:59,260 Then with Colonel passing, Jason industrializes typically no easy task. 19 00:01:59,420 --> 00:02:06,770 So for that reason, I'm going to use an additional utility which is called JQ, and this is a Seelie 20 00:02:06,770 --> 00:02:11,270 tool which enables us to quickly extract information from Jason. 21 00:02:12,660 --> 00:02:18,930 Of course, in order to use JQ, you first have to install it, so using the Alpine package manager, 22 00:02:18,930 --> 00:02:20,340 I'm going to install JQ. 23 00:02:21,640 --> 00:02:27,610 And then in order to extract this information from here, they're going to do is I'm going to pipe this 24 00:02:27,610 --> 00:02:29,740 information to JQ. 25 00:02:32,160 --> 00:02:37,970 And I'm going to specify that I'm interested in the property name, as you have seen in the response 26 00:02:37,970 --> 00:02:41,460 to this, and I'm going to use a flag called the output. 27 00:02:45,720 --> 00:02:53,640 Now, I want to wrap this entire command and perceived output in a variable, so A.W., WCL, I will 28 00:02:53,640 --> 00:03:00,480 give this to you and the response back then, this chaston response will be sent to JQ and will be part 29 00:03:00,480 --> 00:03:02,190 to extract only the domain name. 30 00:03:02,550 --> 00:03:06,930 Then in order to use the domain name in Kerl, we have to set it as a variable. 31 00:03:07,140 --> 00:03:13,590 So the syntax in order to do that is to use SI name because this is the name of the variable that we 32 00:03:13,620 --> 00:03:14,250 are going to use. 33 00:03:14,670 --> 00:03:20,880 It calls the line and between parenthesis we're going to put this entire expression here. 34 00:03:21,480 --> 00:03:24,630 So C name equals dollar and this is the expression. 35 00:03:30,550 --> 00:03:36,490 So the way to curl Comen will look like is following, I'm going to use HTP because we need to specify 36 00:03:36,490 --> 00:03:38,980 the protocol, know how to use the variable name. 37 00:03:40,550 --> 00:03:46,730 And then put that recording, if you look in placement, is actuator, health and actuator info. 38 00:03:47,380 --> 00:03:47,740 OK. 39 00:03:49,800 --> 00:03:53,910 This will be info and the other end point will be health. 40 00:03:58,110 --> 00:04:00,730 Finally, if we run this the way it is, it will still fail. 41 00:04:00,930 --> 00:04:05,910 And the reason for that is when we do update environment, this comment isn't instant. 42 00:04:06,060 --> 00:04:12,210 It takes a few moments for this new environment to be populated with the new version of the application 43 00:04:12,210 --> 00:04:14,790 and for all the services to be restarted. 44 00:04:14,970 --> 00:04:18,839 Everything that we do in the background still takes a few seconds. 45 00:04:19,050 --> 00:04:23,370 I notice that the sweet point for this is around 45 seconds. 46 00:04:23,700 --> 00:04:30,300 So for that reason, I'm going to use an additional sleep here that waits for 45 seconds before doing 47 00:04:30,300 --> 00:04:31,320 the curl comment. 48 00:04:33,750 --> 00:04:41,040 Sleep is not an ideal way to do that, you can write an additional script that may test this every five 49 00:04:41,040 --> 00:04:42,250 seconds or something like that. 50 00:04:42,450 --> 00:04:46,500 So using a sleep will overall make the pipeline a bit slower. 51 00:04:46,500 --> 00:04:49,770 But for our case is a quick and easy way. 52 00:04:49,770 --> 00:04:55,170 So if you're not super concerned about the pipeline, waiting for 45 seconds to check if the deployment 53 00:04:55,170 --> 00:04:57,900 was successful, then this is absolutely no issue. 54 00:04:59,300 --> 00:05:01,550 So it's going to save this and then we'll look at the pipeline. 55 00:05:03,000 --> 00:05:07,530 Now, looking at a console log from the diploid job would be able to see that everything was completed 56 00:05:07,530 --> 00:05:08,170 successfully. 57 00:05:08,230 --> 00:05:12,660 So we have deployed version number 10 and then we look at the endpoint point. 58 00:05:12,660 --> 00:05:16,400 You would be able to see that this is version number 10 and the status is up. 59 00:05:16,860 --> 00:05:18,390 Everything seems to be working fine. 60 00:05:18,840 --> 00:05:22,530 When you're writing any kind of tests, you always want to make sure that they fail. 61 00:05:22,830 --> 00:05:27,540 So for that reason, I invite you to reduce the sleep or to remove it altogether so that you can see 62 00:05:27,540 --> 00:05:31,350 the error when the version is different from what you would have expected.