Changelog

NGINX migration plan update, MCP updates, AI Copilot Deployment troubleshoot, Build dashboard

Hey Team,

This release focuses on a critical infrastructure transition, along with new ways to troubleshoot deployments faster and better integrate Qovery with AI workflows.

Let’s start with an important update you should be aware of.

⚠️ NGINX Ingress to API Gateway migration - Plan Update

The NGINX Ingress Controller repository has now been officially archived. This confirms that the transition away from it is no longer optional. It is something every team needs to plan for.

We have updated our migration plan to give you more time to validate your services on the new Envoy Gateway stack while keeping the rollout moving forward.

👉 Full migration plan available here

Updated rollout timeline

  • May 4, 2026: Envoy becomes the default on non production clusters
  • May 18, 2026: Envoy becomes the default on production clusters

We strongly recommend testing your applications on the new Envoy based routing layer before these dates to ensure everything behaves as expected.

As always, Qovery handles the migration, but validation on your side is key for a smooth transition.

🤖 Troubleshoot deployments with the AI Copilot

You can now use the Qovery AI Copilot to troubleshoot deployment failures automatically.

Instead of manually digging through logs and events, the Copilot analyzes your deployment logs, application logs, and deployment history to identify the root cause and suggest a resolution.

This helps teams:

  • Resolve incidents faster
  • Reduce time spent debugging complex failures
  • Remove friction from day to day operations

The feature is available for all users with Copilot enabled and is currently free while we explore usage and gather feedback. It remains in beta and improvements are already in progress.

🔌 MCP updates

We reworked our MCP server to better align with standard MCP patterns.

Instead of a single entry point, MCP now exposes multiple tools depending on the use case. This makes integrations with AI agents, especially coding assistants like Claude, more predictable and powerful.

We also introduced OAuth authentication. You can now connect using your Qovery account without relying on static tokens.

MCP is available to everyone at no cost and remains our standard interface to interact with Qovery programmatically.

We are just getting started. More tools will be added progressively.

🏗️  Get build insights via the Build Dashboard

Understanding what happens during a build is often a black box. When builds are slow or failing, it is not always clear whether the issue comes from your code or from resource limitations.

We introduced a Build Dashboard that gives you visibility into the resources used during the build process, including CPU, memory, and network usage.

This helps you:

  • Identify bottlenecks during builds
  • Understand if your builders are under provisioned
  • Improve your docker file in case of network limiting factors

With this level of insight, you can optimize build performance instead of guessing what is going wrong.

🛠️ Minor updates

  • Terraform provider documentation improvements: We improved the Terraform provider documentation to make it more complete and easier to use. You will now find clearer explanations, missing details filled in, and more practical examples to help you get started faster or go deeper in advanced use cases.

That is it for this release. This one is especially important as we move closer to the Envoy migration, so make sure to review the plan and test your workloads.

Talk soon,

The Qovery Team 🚀

Alessandro Carrano
Head of Product
Share on :
Twitter icon
linkedin icon

It’s time to change
the way you manage K8s

Turn Kubernetes into your strategic advantage with Qovery, automating the heavy lifting while you stay in control.