nSights Talks

Datadog Synthetic Monitoring

Tutorial Highlights & Transcript

00:00 - Beginning of Video
Today I’m going to discuss Datadog Synthetic Monitoring. Let me start it.
00:14 - What is Synthetic Monitoring?
Synthetic tests are proactive, automated testing that ensures the availability, and the performance of the business transactions and endpoints. Synthetic tests are difficult to create and maintain at scale. Scripting synthetic tests requires coding skills and running out requires dedicated infrastructure. In Datadog, it allows us to create easier for us. Datadog synthetic monitoring provides code-free synthetic monitoring. It also provides very easy user interfaces. Datadog synthetic monitoring allows us to create code-free tests that proactively simulate user transactions on our application and monitor key network endpoints across various layers of our systems. There are two types of monitoring and synthetic monitoring.
01:16 - Browser Tests
The first one is browser tests, and another one is the API test. Browser tests are the scenarios that Datadogs executes on our web application. You can configure periodical intervals to run tests from multiple locations, devices and processes as well as execute them for our CI/CD pipelines also. These tests verify that users can perform key business transactions on our application, and that they are not negatively impacted by the most recent code changes. Datadog’s new automated browser test enables us to automate user experience monitoring and ensures the user can complete action-like signing up for an account or adding the things in the cart, and these kinds of stuff can be monitored, or anyone on our team can record and automate multi-stemmed browser test within a minute so that it is easier to configure things. Only users can create a test. Datadog user machine learning detects changes to the application and automatically updates our test accordingly. Datadog browser test provides us the end-to-end visibility for troubleshooting issues like when a browser test fails because of the front-end or back-end issues in our application, Datadog provides the context to troubleshoot the issues quickly. Screenshot from the text shows you what users are seeing before the key element disappears from the pages due to some bugs or some server issues and all
03:20 - API Tests
API tests proactively monitors their most important servers and are available at any time and from anywhere else. Single API test comes from different layers of– A single API test allows us to launch requests on different networks, layers of the systems like HTTP, SSL, DNS, TCP and ICMP. Multi API tests enables us to run HTTP tests in sequence to monitor the uptime of the key journey at the API levels. Now, I’m going to demonstrate Datadog synthetic monitoring.
04:09 - Demo of Datadog Synthetic Monitoring

On the Datadog page, locate UX monitoring. Under that, I’m going to create a new test. On that, I’m going to select an API test. In this API test, we can choose multiple layers of networks like HTTP, SSL, TCP and DNS like that. I’m going to select HTTP. Before that, I’m going to paste the URL www.nclouds.com. You can mention the environment if required, I’m going to leave it. If I test this URL, there is an assertion automatically created like this response time, status code, headers, and we can be able to customize the assertions also according to our vision. I’m going to leave it as this as a default. I’m going to select a location. I’m selecting one location from each. We can also specify time frequencies according to our wish. I’m going to leave it for one minute. Define alert condition. I’m going to leave it as a default thing. If we can also acknowledge a team member, or we can notify it by SNS and those kinds of things. I’m going to leave it as this, and I’m going to create the test, and here we can monitor the response time by location, network timing and the test results. Just created a status check, so it displaces, and we can also create multiple API tests. The difference between the single and the multiple API test is, here we can be able to monitor multiple applications status in a single API check. We can only monitor the HTTP request only. I’m going to give the name of this as DEMO. Environment, I’m going to leave it. I’m selecting only one region, so Mumbai. Create your first step. I’m going to name it DEMO. Paste the URL. Once I test, the assertions are created automatically. I’m going to save this. Here we can add another request under the same AP test. This is the difference between the single API and multiple API tests. Here I’m going to paste this URL. Name it as DEMO-2, and it’s response is 200. Everything seems fine. This is how we can be able to manage multiple applications under the same API test. Now, I’m going to explain about the browser test here. Here we can be able to monitor the user’s browser experience. For that, I’m going to paste this URL here and name it as DEMO. For this environment, I’m going to leave this. Additional tags if we’re required, we can select. Location: I’m going to select Mumbai. Specific test frequency. Minutes. Obviously, we should want at least five minutes. Then I’m going to record this, test and record. Once I started record this, for example, if I want to create a registry, Register Now, first if I selected any options, it will be recorded here so that we can able to monitor our application performance in the user’s perspective so we can able to troubleshoot the issue before the issue arrives from the user side. This is the advantages of the Synthetic Monitoring, and that’s it for this demo.

Jasmeet Singh

Vignesh Selvaraj

Support Engineer


Vignesh joined nClouds in 2021 as a Support Engineer. He helps build infrastructure in AWS with his experience in AWS, DevOps, and cloud computing.