Setting up and running availability tests in Azure DevOps

Setting up and running availability tests in Azure DevOps

Setting up and running availability tests in Azure DevOps involves creating health endpoints in your application and using tools to monitor their availability and responsiveness.

Here’s a step-by-step guide.

1. Create Health Endpoints in Your Application

Health endpoints are essential for monitoring the status of your application. They typically expose critical information about the application's health.

Example: Health Endpoint

1. ASP.NET Core:

   • Use the Microsoft.Extensions.Diagnostics.HealthChecks package.

   • Add the following code in Startup.cs or Program.cs:

     xxxxxxxxxx
     12
     1
     public void ConfigureServices(IServiceCollection services)
     2
     {
     3
        services.AddHealthChecks();
     4
     }
     5
     public void Configure(IApplicationBuilder app, IWebHostEnvironment env)
     6
     {
     7
        app.UseRouting();
     8
        app.UseEndpoints(endpoints =>
     9
        {
     10
           endpoints.MapHealthChecks("/health");
     11
        });
     12
     }
     

   • This creates a /health endpoint that returns HTTP 200 if the application is healthy.

2. Node.js: Use the express framework:

   xxxxxxxxxx
   8
   1
   const express = require('express');
   2
   const app = express();
   3
   app.get('/health', (req, res) => {
   4
      res.status(200).send('OK');
   5
   });
   6
   app.listen(3000, () => {
   7
      console.log('Health check endpoint running on port 3000');
   8
   });
   

3. Java: Use the Spring Boot Actuator library:

   xxxxxxxxxx
   7
   1
   @RestController
   2
   public class HealthController {
   3
      @GetMapping("/health")
   4
      public ResponseEntity<String> health() {
   5
         return new ResponseEntity<>("OK", HttpStatus.OK);
   6
      }
   7
   }
   

2. Use Tools to Test Health Endpoints

Azure DevOps and external tools can monitor health endpoints to ensure application availability.

Azure Monitor Application Insights

1. Set Up Application Insights:

   • In the Azure Portal, create an Application Insights resource.

   • Configure your application to send telemetry to Application Insights.

2. Enable Availability Monitoring:

   • Go to your Application Insights resource.

   • Navigate to Availability > Add Test.

3. Types of Availability Tests

3.1 URL Ping Test

A URL ping test ensures your endpoint is reachable and responsive.

Steps to Create a URL Ping Test:

1. Go to Application Insights > Availability > Add Test.

2. Select URL Ping Test.

3. Fill in the details:

   • Name: A descriptive name for the test.

   • URL: The endpoint to test (e.g., https://yourapp.com/health).

   • Test Frequency: How often the test runs (e.g., every 5 minutes).

   • Test Locations: Select Azure regions to test from.

4. Enable alerts for failures:

   • Configure email, SMS, or webhook notifications for failures.

3.2 Multi-Step Web Test

A multi-step web test checks the functionality of a sequence of interactions with your application (e.g., login, search, checkout).

Steps to Create a Multi-Step Web Test:

1. Write a Kusto Query or use a tool like Visual Studio Web Test to simulate multiple steps:

   • Example for a simple two-step test:

     • Step 1: Access the homepage.

     • Step 2: Check the health endpoint.

2. Upload the test script to Application Insights:

   • Go to Availability > Add Test.

   • Select Multi-Step Web Test and upload the script.

3. Configure test frequency and locations as with the URL Ping Test.

4. Automate Availability Tests in Azure DevOps

4.1 Monitor Health Endpoints

Add a step to your Azure DevOps pipeline to check health endpoints.

Example YAML for URL Health Check:

xxxxxxxxxx
8
1
trigger:
2
- main
3
pool:
4
  vmImage: 'ubuntu-latest'
5
steps:
6
- script: |
7
    curl -f https://yourapp.com/health || exit 1
8
  displayName: 'Check Health Endpoint'

If the health endpoint fails, the pipeline will stop.

4.2 Run Multi-Step Web Tests in Pipelines

Use tools like Selenium or Playwright for multi-step availability tests.

Example YAML for Playwright:

xxxxxxxxxx
8
1
trigger:
2
- main
3
pool:
4
  vmImage: 'ubuntu-latest'
5
steps:
6
- script: |
7
    npx playwright test availability-tests.spec.js
8
  displayName: 'Run Multi-Step Web Test'

5. Monitor and Report Results

1. Use Azure Monitor Alerts to notify stakeholders of failures.

2. Analyze results in Application Insights:

   • Metrics: View test success rate, latency, and failure reasons.

   • Logs: Use KQL to query test results for deeper analysis.

Summary

By combining health endpoints, Azure Application Insights, and automated pipelines, you can ensure your application's availability and proactively address issues. Let me know through your comments, if you'd like code examples or detailed guidance for specific tools.