Takeaway: Azure Chaos Studio helps organizations validate application resilience by injecting controlled faults that simulate outages, failovers, network disruptions, and infrastructure failures before they impact production.
A resilient architecture is a good starting point. But how do you know your application will keep working when a dependency fails, a zone becomes unavailable, or a Kubernetes pod stops responding? You need to test what happens, not just assume the design will hold.
Today, we’re sharing two updates to Azure Chaos Studio: Chaos Studio Workspaces is now generally available, and the Chaos Studio extension for Azure Kubernetes Service (AKS) is now in public preview. Together, they make it easier to test realistic failure scenarios and understand how those failures affect your application.
Workspaces: Start with an outage, not a list of faults
With Chaos Studio Workspaces, you can choose the application or Azure resources you want to test, then select from scenarios based on real failure patterns. Workspaces discovers resources in your chosen scope and recommends relevant scenarios, such as compute zone down, a simulated DNS outage, database failover, or cache stampede. You can also tailor scenarios to your needs. Each scenario injects controlled faults that simulate selected effects of these outages on supported resources in your scope, not an actual Azure zone or service outage. Compute zone down, for example, shuts down the virtual machines and scale set instances in one availability zone, and the Microsoft Entra ID outage scenario blocks network access from your resources to Microsoft Entra ID endpoints.
Instead of building every test from individual faults, you can focus on the question that matters: will your application continue serving users, and will it recover as expected? Each run produces an execution report you can share with your engineering, operations, and review teams. Pair it with your application health checks and monitoring to assess availability and recovery. You can also bring these tests into development workflows through APIs, automation, and supported AI-assisted tools.
AKS extension: Test what happens inside your cluster
For applications running on AKS, infrastructure-level testing is only part of the picture. A cluster may replace a failed pod while users still experience dropped requests, slow responses, or lost transactions.
The new Chaos Studio extension for AKS lets you inject controlled pod-level faults from Workspaces, including pod deletion, CPU and memory pressure, and network latency or loss. That means you can test whether your services maintain availability, whether scaling and recovery behave as expected, and whether your retry and timeout settings protect the end-user experience. The extension uses LitmusChaos under the hood, without requiring you to author Litmus resources yourself.
Get started
If you’re new to chaos engineering, start with a focused test in a non-production environment, monitor the application experience, and use the results to close any gaps. Explore the Azure Chaos Studio product page, read our earlier Azure blog on proving application resilience with Chaos Studio, and watch the Workspaces walkthrough on YouTube. For more background on resilience testing, see Advancing Microsoft Azure resilience with Chaos Studio.
Related resources
- Azure Chaos Studio
- Microsoft Azure Well-Architected Framework — Reliability
- Recommendations for using availability zones and regions
- Business continuity and disaster recovery
- Reliability guides for Azure services
- Chaos Studio Workspaces documentation
- Scenario catalog
- Quickstart: create a Workspace and run a Scenario
- Announcing Azure Infrastructure Resiliency Manager Public Preview | Microsoft Community Hub