Announcing CI/CD Enhancements for Azure Load Testing

We are excited to announce a significant update to our Azure Load Testing service, aimed at enhancing the experience of setting up and running load tests from CI/CD systems, including Azure DevOps and GitHub. This update is a direct response to customer feedback and is designed to streamline the process, making it more efficient and user-friendly.

Key Features and Improvements:

1. Enhanced CI/CD Integration: Developers and testers can now configure application components and the metrics to monitor directly from a CI/CD pipeline. This integration allows monitoring the application infrastructure during the test run. You can make the following changes to your load test YAML config.

   appComponents:
     - resourceId: "/subscriptions/00000000-0000-0000-0000-000000000000/resourceGroups/samplerg/providers/microsoft.insights/components/appComponentResource"
       resourceName: appComponentResource #Optional
       kind: web # Optional
       metrics:
         - name: "requests/duration"
           namespace: microsoft.insights/components 
           aggregation: "Average"
         - name: "requests/count"
           aggregation: "Total"
           namespace: microsoft.insights/components   
     - resourceId: "/subscriptions/00000000-0000-0000-0000-000000000000/resourceGroups/samplerg/providers/microsoft.insights/components/appComponentResource"
       resourceName: appComponentResource #Optional
       kind: web # Optional
       metrics:
         - name: "requests/duration"
           aggregation: "Average"
           namespace: microsoft.insights/components
         - name: "requests/count"
           aggregation: "Total"
           namespace: microsoft.insights/components

2. Pass/Fail Criteria on Server Metrics: Users can set pass/fail criteria on server metrics from a CI/CD pipeline, providing more granular control over test outcomes. This feature helps in maintaining high performance standards by automatically flagging any performance issues. You can make the following changes to your load test YAML config. 

   failureCriteria:
     clientMetrics:
       - avg(responseTimeMs) > 300
       - percentage(error) > 50
       - getCustomerDetails: avg(latency) > 200
     serverMetrics:
       - resourceId: /subscriptions/abcdef01-2345-6789-0abc-def012345678/resourceGroups/sample-rg/providers/Microsoft.Compute/virtualMachines/sample-vm
         metricNamespace: Microsoft.Compute/virtualMachines
         metricName: Percentage CPU
         aggregation: Average
         condition: GreaterThan
         value: 80
       - resourceId: /subscriptions/abcdef01-2345-6789-0abc-def012345678/resourceGroups/sample-rg/providers/Microsoft.Compute/virtualMachines/sample-vm
         metricNamespace: Microsoft.Compute/virtualMachines
         metricName: Available Memory
         aggregation: Average
         condition: LessThan
         value: 20

    

3. Parameter Overrides: The ability to override parameters of a load test configuration YAML from the Azure DevOps task or GitHub action adds flexibility and customization to the testing process.
4. Output Variables: The Azure DevOps task now includes output variables that can be consumed in downstream steps, jobs, and stages. This feature to take further actions on the load test results within the pipeline.
5. Pipeline Cancellation: If a pipeline in Azure Pipelines or a workflow in GitHub is cancelled, any load test triggered by the pipeline/action will also be cancelled. This ensures avoiding costs for unnecessary tests.
6. Traceability and Results Viewing: Users can trace a test run back to the pipeline that ran the test from Azure portal. This helps in end-to-end traceability to understand what changes might have triggered the test failure.

Conclusion

These enhancements are designed to provide a more integrated and efficient load testing experience for our users. We believe that these updates will help developers, testers, and DevOps engineers to better manage their load testing processes, ensuring high performance and reliability of their applications. 

We look forward to your feedback and are excited to see how these new features will improve your CI/CD workflows. Stay tuned for more updates and happy testing!