1 00:00:00,450 --> 00:00:01,680 A lot don't come back. 2 00:00:01,710 --> 00:00:07,470 Let's talk about monitoring and logging in elastics beanstalk and the patterns and type patterns for 3 00:00:07,510 --> 00:00:10,430 elastic beanstalk elastic beanstalk and cloud watch. 4 00:00:10,440 --> 00:00:15,600 So that's being stopped automatically we use Amazon Cloud watch to help you monitor your application 5 00:00:15,630 --> 00:00:16,530 and environments. 6 00:00:16,530 --> 00:00:21,300 So basically what are saying is being beanstalk is integrated with cloud which you can define custom 7 00:00:21,300 --> 00:00:22,010 metrics. 8 00:00:22,020 --> 00:00:27,330 Also not only the standard ones for your own use an elastic beanstalk will push those metrics to Amazon 9 00:00:27,360 --> 00:00:32,510 Cloud watch once you define the metrics you can view them in the Amazon cloud watch console. 10 00:00:32,580 --> 00:00:37,980 You can also leverage Amazon cloud or alarms to help you implement decisions more easily by enabling 11 00:00:37,980 --> 00:00:43,080 you to send notification automatically make changes like the alarms in cloud which I don't need to elaborate 12 00:00:43,080 --> 00:00:49,740 on that but it helps you take actions based on specific events that could happen within your environment. 13 00:00:49,750 --> 00:00:56,970 Last week when stock and cloud watch logs with cloud watch logs you can monitor and archive in cloud 14 00:00:56,970 --> 00:01:01,650 watch logs itself which saves it indefinitely unless you're configured to do otherwise or it can even 15 00:01:01,920 --> 00:01:09,150 send them to as three an archive your Lusignan took applications system custom log files from an easy 16 00:01:09,150 --> 00:01:11,160 to instances of you environment. 17 00:01:11,160 --> 00:01:17,250 You can also configure alarms that make it easier for you to react to a specific log stream events that 18 00:01:17,260 --> 00:01:18,750 your metric filters extract. 19 00:01:18,750 --> 00:01:23,910 So basically the cloud watch logs are going to send the logs from the servers from being stuck to a 20 00:01:23,910 --> 00:01:25,580 specific log group. 21 00:01:25,620 --> 00:01:31,610 And in that log group you can apply filters and based on the filters you can make alarms and based on 22 00:01:31,620 --> 00:01:37,500 alarms you do actions or react to certain events which logs always been unable to do what you love no 23 00:01:37,500 --> 00:01:44,250 matter what is going to install a log agent when it each instance in your environment which publishes 24 00:01:44,250 --> 00:01:50,280 so that logs agent is what publishes the meta data points or the custom metrics to cloud service for 25 00:01:50,310 --> 00:01:57,090 each log group you configure each log group from the cloud that your group applies its own filter patterns 26 00:01:57,090 --> 00:02:04,140 to determine what logs team events to send to cloud watch as data points or as metrics as values for 27 00:02:04,200 --> 00:02:10,080 metrics and these logs teams that belong to the same group will share the same retention monitoring 28 00:02:10,140 --> 00:02:12,090 and access control settings. 29 00:02:12,120 --> 00:02:18,630 You can configure elastic beanstalk to automatically stream logs to the cloud watch service in addition 30 00:02:18,630 --> 00:02:24,660 to instant logs What else can I monitor with the watch logs if you enable Enhanced Health. 31 00:02:24,660 --> 00:02:28,620 So this is a differentiator for lastic beanstalk that works. 32 00:02:28,950 --> 00:02:35,130 Enhanced Health for our environment so it looks at the web server logs it looks at the code logs it 33 00:02:35,130 --> 00:02:38,750 looks at the application errors if you want certain errors. 34 00:02:38,760 --> 00:02:41,390 All that can be monitored by cloud watch logs. 35 00:02:41,550 --> 00:02:46,050 You can configure that environment to stream health information to cloud watch log so its not only that 36 00:02:46,050 --> 00:02:52,050 easy to instances but everything thats why we are saying when it comes to application the best is buying 37 00:02:52,050 --> 00:02:55,060 stock from a visibility monitoring and automation perspective. 38 00:02:55,080 --> 00:03:02,850 Things in the middle and cloud formation is the fire and the last one to provide detailed health information 39 00:03:02,850 --> 00:03:05,550 about easy to instances running in your environment. 40 00:03:05,550 --> 00:03:07,980 Elastic Winstock includes a health agent. 41 00:03:08,040 --> 00:03:14,070 Now we have a cloud watch an agent and a health agent in the am right for each platform configuration 42 00:03:14,070 --> 00:03:15,700 that supports Enhanced Health. 43 00:03:15,710 --> 00:03:20,760 Remember when we were talking about the custom platforms and I told you that the lies that Winstock 44 00:03:20,760 --> 00:03:26,640 has built for each different platform for each different programming language already have some customization. 45 00:03:26,640 --> 00:03:27,970 Part of that is the health. 46 00:03:27,990 --> 00:03:34,140 So if you are to customize one of the being stuck in mice to build a custom EMI with other softwares 47 00:03:34,170 --> 00:03:40,920 that you need make sure that you are maintaining what it has done in them for health monitoring to maintain 48 00:03:41,040 --> 00:03:46,050 that ability or that capability the health agent monitors with server blocks. 49 00:03:46,050 --> 00:03:50,920 So we're looking now at the web server logs system metrics and relate them to elastic beanstalks service 50 00:03:51,000 --> 00:03:56,310 elastic beanstalk will analyze the metrics along with data from elastic load balancing from an easy 51 00:03:56,310 --> 00:04:00,090 to of scaling and provide an overall picture of the environment health. 52 00:04:00,090 --> 00:04:05,310 So now if you see that what you're seeing ministroke is doing is looking at the software logs elastic 53 00:04:05,370 --> 00:04:06,180 load balancer. 54 00:04:06,260 --> 00:04:10,460 Look it's looking at all the scaling is it for instance is looking at all of that and analyzing that 55 00:04:10,800 --> 00:04:16,290 and it provides on the concert on the dashboard you can see that your application is green OK with a 56 00:04:16,300 --> 00:04:22,380 checkmark So you are good and it can provide different colors based on the health of your environment 57 00:04:22,500 --> 00:04:25,890 based on the analysis that it has done on your behalf. 58 00:04:25,890 --> 00:04:30,960 Now let's look at exporting logs to S-3 from cloud watch logs. 59 00:04:30,960 --> 00:04:38,310 You can export log data that has been sent from beanstalk to cloud watch logs to an Amazon S3 bucket 60 00:04:38,400 --> 00:04:45,350 and use this data in a custom processing analysis or to load it into other systems if in a scenario 61 00:04:45,350 --> 00:04:52,230 in the exam part of plastic bins stock options where that you retain the logs to an S3 bucket that can 62 00:04:52,230 --> 00:04:56,910 happen if you are told that it's going to be sent to clog watch logs and cloud watch will be configured 63 00:04:56,910 --> 00:04:58,980 to send that to an S3 bucket of your choice. 64 00:04:58,980 --> 00:05:00,570 That is also possible. 65 00:05:00,720 --> 00:05:01,280 OK. 66 00:05:01,420 --> 00:05:05,340 But in AABs works do you have the option to send the logs directly to a street. 67 00:05:05,340 --> 00:05:05,790 No. 68 00:05:05,830 --> 00:05:11,260 You have to send it to club watch logs and configure clock watch looks to archive it to s three to begin 69 00:05:11,260 --> 00:05:12,340 the export process. 70 00:05:12,340 --> 00:05:15,830 You must create an three bucket to store the exported log data. 71 00:05:15,850 --> 00:05:19,060 So it's not the same bucket that elastic Winstock would create. 72 00:05:19,060 --> 00:05:20,090 It's a different one. 73 00:05:20,110 --> 00:05:25,270 In this case because we're talking about cloud watch logs sending to three. 74 00:05:25,540 --> 00:05:32,140 You can store the exported files in your Amazon S3 bucket and define Amazon as the lifecycle to archive 75 00:05:32,140 --> 00:05:38,650 that to Glacier automatically and you can export the logs from multiple groups or multiple time ranges 76 00:05:38,880 --> 00:05:41,170 to the same as three bucket. 77 00:05:41,230 --> 00:05:43,970 And if you want to separate the log data for each task. 78 00:05:44,020 --> 00:05:51,070 This is more of this ops but you can have a Prefect's to each group and that Prefect's will allow you 79 00:05:51,070 --> 00:05:55,180 to differentiate the logs in the common S-3 bucket. 80 00:05:55,180 --> 00:05:58,590 Now let's look at logging where the docker platform configurations. 81 00:05:58,600 --> 00:06:06,880 Can I do similar if I have used elastic beanstalk to deploy my application on a dock or container platform 82 00:06:06,970 --> 00:06:08,030 looking with docker. 83 00:06:08,260 --> 00:06:13,440 So specify that directory inside the container to which your application writes locks. 84 00:06:13,450 --> 00:06:19,750 So now in your application you are instructed to apply the logs to a specific folder and then you can 85 00:06:19,990 --> 00:06:26,500 configure the doctor to send the slug's or a certain logging folder that will be rotated saved locally 86 00:06:26,500 --> 00:06:32,890 archived locally and then rotated or sent to the cloud trail elastic beanstalk is integrated with cloud 87 00:06:32,890 --> 00:06:38,420 trail and it will send all the AP eyes as the other services are all you need to do is you need to configure 88 00:06:38,440 --> 00:06:44,420 a trail and then all the logs of the API calls is going to be coming into your trail. 89 00:06:44,530 --> 00:06:49,030 OK let's hop through the patterns and anti-protons and I promise you this is going to be the last thing 90 00:06:49,060 --> 00:06:54,190 because I know elastic beanstalk we probably have spent a lot of time on it but it is worth it because 91 00:06:54,850 --> 00:06:59,680 they use dark or they use logging and all that in a scenario and you are completely lost especially 92 00:07:00,010 --> 00:07:06,870 with that gray area between Officeworks beanstalk and cloud formation that I hope we have already clarified 93 00:07:06,870 --> 00:07:12,970 now and there is no ambiguity about the fine line differences between the different services. 94 00:07:13,100 --> 00:07:18,070 So patterns when we use plastic bins or when we had advised advisement to use block mid-stroke if you 95 00:07:18,070 --> 00:07:20,970 want to quickly deploy an application prototype for testing. 96 00:07:21,160 --> 00:07:26,070 I don't know anything about it and I don't know anything about it as then just know enough about the 97 00:07:26,130 --> 00:07:31,330 Winstock deployer application no or minimal knowledge about it its infrastructure when deploying an 98 00:07:31,330 --> 00:07:32,260 application to it. 99 00:07:32,270 --> 00:07:34,090 Yes that's a good use case. 100 00:07:34,330 --> 00:07:39,730 Migrating with applications that are running on custom application servers on premise to it. 101 00:07:39,800 --> 00:07:40,260 Yes. 102 00:07:40,270 --> 00:07:45,580 So what do I need to do if I'm going to use Dharker you create a file include all its components and 103 00:07:45,580 --> 00:07:52,000 dependencies create an image of the local file upload the image to a public or private dock or image 104 00:07:52,000 --> 00:07:54,600 repository and deploy it with elastic beanstalk. 105 00:07:54,610 --> 00:08:00,910 And of course you can add here if you wanted to that you can add here or rotate the logs to a tree from 106 00:08:01,080 --> 00:08:03,270 the environment directly. 107 00:08:03,300 --> 00:08:04,720 A software development project. 108 00:08:04,720 --> 00:08:08,560 So that's another good use case you have some friendly building project where application developers 109 00:08:08,560 --> 00:08:12,570 could be working on different compute with different operating systems. 110 00:08:12,610 --> 00:08:18,160 So what they can do is you can abstract the application code from the underlying compute or else by 111 00:08:18,160 --> 00:08:25,270 creating an identical environment by creating a darker image of the application and send it to all developers 112 00:08:25,270 --> 00:08:29,170 to basically create a file of the application send it to all developers. 113 00:08:29,170 --> 00:08:35,320 They can develop based on that and the Beanstalk anti-pattern when not to use that first one if you 114 00:08:35,320 --> 00:08:40,600 know exactly and you have the skills and resources and you know exactly what services you want to use 115 00:08:40,630 --> 00:08:45,370 and how they will work and how they need to be configured and everything and you would like to have 116 00:08:45,370 --> 00:08:48,910 a lot of control then you can go to cloud formation. 117 00:08:48,910 --> 00:08:50,740 In that case if you know everything about it. 118 00:08:50,820 --> 00:08:56,260 You have complete control over the resource configuration is required then cloud formation or assault 119 00:08:56,290 --> 00:08:57,340 will be the way. 120 00:08:57,650 --> 00:08:59,620 Right so that was about elastic beanstalk. 121 00:08:59,620 --> 00:09:05,650 Let's have a well-deserved break now and I will see you in another lecture with another interesting 122 00:09:05,800 --> 00:09:07,090 topic from yours. 123 00:09:07,120 --> 00:09:07,570 Thank you.