Migrate to Kubernetes

Overview

Move off Heroku, Render, or Railway onto infrastructure you own. The migration runs alongside your current platform and does not require months of planning, rewriting deployment configs, learning Helm charts, or re-architecting your stack. Qovery removes that complexity entirely. One command: an AI agent reads your repository and deploys it into the cloud account you already own (AWS, GCP, Azure, or Scaleway), after you approve the plan. It analyzes your codebase, generates Dockerfiles, provisions databases, sets up environment variables, and deploys everything as standard Terraform and Kubernetes manifests. No re-architecture, no Kubernetes expertise required. The migration is non-destructive: your new environment runs alongside your current platform. Switch your domain over only once you’re confident. Stop paying for your old PaaS and everything keeps running, on infrastructure and cloud billing you fully own.

For complex migrations (microservices architectures, stateful workloads, compliance requirements), Qovery provides cloud architect assistance. Talk to a migration engineer to get a personalized migration plan.

Why Migrate with Qovery?

AI-Powered Migration

The /qovery-onboard AI skill maps your current platform’s concepts to Qovery and guides your cluster setup, then qovery-deploy analyzes your codebase, creates Dockerfiles, provisions databases, and deploys - all from a single prompt

No Kubernetes Expertise

Qovery abstracts the complexity of Kubernetes. You deploy applications - Qovery handles pods, ingress, TLS, RBAC, and networking

Your Cloud, Your Control

Everything runs on your own AWS, GCP, Azure, or Scaleway account. No vendor lock-in, no shared infrastructure

Cloud Architect Assistance

Qovery’s team helps with complex migrations. Book a demo to get a personalized migration plan

Compliance Ready

SOC2 Type II, ISO 27001, DORA, HIPAA, GDPR compliant infrastructure from day one

Cost Savings

Right-sized from the start. No over-provisioning, no surprise bills. Customers report $200K+ annual savings vs. DIY Kubernetes

Migrate from Any Platform

Heroku

Dynos, add-ons, Heroku Postgres, Redis - all mapped to Kubernetes equivalents automatically

Render

Web services, background workers, cron jobs, managed databases - migrated with full parity

Railway

Services, plugins, and managed databases - mapped to Kubernetes equivalents automatically

OpenShift

DeploymentConfigs, Routes, BuildConfigs - converted to standard Kubernetes resources

VMs / Bare Metal

Legacy applications running on EC2, DigitalOcean droplets, or bare metal servers

Docker Compose

Multi-container setups defined in docker-compose.yml - each service becomes a Qovery application

Other PaaS

Fly.io, DigitalOcean App Platform, Google App Engine, Azure App Service, and more

How It Works

Install the Qovery AI skill (one command)
Ask the AI agent to migrate your project
AI analyzes your codebase - detects languages, frameworks, databases, dependencies
AI generates optimized Dockerfiles if missing
AI provisions databases (PostgreSQL, Redis, MongoDB, MySQL) and configures environment variables
AI deploys everything to Kubernetes on your cloud provider
You verify and go live

Getting Started

Option 1: AI-Powered Migration (Recommended)

The fastest path. The AI agent handles the entire migration end-to-end.

Install the Qovery AI Skill

One command installs the skill globally for all your projects:

curl -fsSL https://skill.qovery.com/install.sh | bash

Open Your AI Coding Tool

Launch OpenCode, Claude Code, Cursor, VS Code Copilot, or any compatible AI coding tool.

Ask the Agent to Migrate

Migrate me from Heroku to AWS

Or use the skill command directly:

/qovery-onboard

The qovery-onboard skill acts as your personal cloud architect:

Understand your context - platform you’re leaving, cloud provider you’re moving to, constraints
Map concepts from your current platform (dynos, add-ons, services…) to their Qovery equivalents
Guide you through cluster and environment setup

From there, qovery-deploy takes over to get your code running:

Analyze your codebase and detect languages, frameworks, and databases
Create optimized multi-stage Dockerfiles if your project doesn’t have one
Provision databases (PostgreSQL, Redis, MongoDB, MySQL) - managed or containerized
Set up environment variables and secrets
Deploy everything to Kubernetes on your cloud provider
Watch the deployment and auto-fix any issues (up to 3 retries per service)

The AI agent supports Node.js, Python, Go, Java, Ruby, PHP, .NET, React, Vite, Next.js, Rails, Django, Spring, Laravel, and many more frameworks.

Migrate Environment Variables

The AI agent handles environment variable migration as part of the deployment process.For Heroku specifically, you can also export and import manually using the Qovery CLI:

qovery auth
qovery context set
heroku config --app <your_heroku_app_name> --json | \
  qovery env parse --heroku-json > heroku.env && \
  qovery env import heroku.env && \
  rm heroku.env

Import sensitive data (API keys, credentials, tokens) as Secrets, not Environment Variables.

Verify and Go Live

Once the deployment is complete:

Check your application status in the Qovery Console
Test your endpoints using the auto-generated Qovery URL
Configure your custom domain when ready
Update your DNS records to point to Qovery

Option 2: Manual Migration

For teams that prefer full control over every step.

Create Your Dockerfile

If your project doesn’t have a Dockerfile, create one. Qovery provides templates for all major frameworks.Example for a Node.js application:

FROM node:20-alpine AS builder
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build

FROM node:20-alpine
WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/node_modules ./node_modules
EXPOSE 3000
CMD ["node", "dist/index.js"]

Use multi-stage builds to keep your images small and secure. The AI agent generates these automatically when you use Option 1.

Create Resources on Qovery

Sign in to the Qovery Console
Create your Organization and Project
Create your production Environment
Add your application (select your Git repository, branch, and port)
Add your database (PostgreSQL, MySQL, MongoDB, or Redis - managed or containerized)

Configure Environment Variables

Export your environment variables from your current platform
Import them into Qovery via the Console or CLI
Create environment variable aliases to connect your services (e.g., link your app to your database using DATABASE_URL)

Deploy

Click “Deploy” in the Qovery Console or use the CLI:

qovery deploy

Need Help? Talk to a Cloud Architect

Book a Migration Demo

For complex migrations - microservices, stateful workloads, compliance requirements, or large-scale infrastructure - Qovery provides personalized cloud architect assistance. Get a migration plan tailored to your stack.

What Gets Migrated?

Source	Qovery Equivalent	Handled by AI?
Web applications (any language/framework)	Applications - containerized on Kubernetes	Yes
Background workers (Sidekiq, Celery, etc.)	Applications - separate services	Yes
Cron jobs	Cron Jobs	Yes
PostgreSQL, MySQL	Managed Database (AWS RDS, GCP Cloud SQL, etc.)	Yes
Redis, MongoDB	Container Database or Managed	Yes
Environment variables	Environment Variables & Secrets	Yes
Custom domains	Custom Domains	Manual DNS update required
Add-ons (Memcached, Elasticsearch, etc.)	Helm Charts or container services	Partially
Scheduled tasks	Lifecycle Jobs	Yes

FAQ

How long does migration take?

For most applications, less than a day. The AI agent handles Dockerfile creation, database provisioning, environment variable setup, and deployment in minutes. Complex multi-service architectures may take longer - book a demo for a personalized timeline.

Do I need Kubernetes expertise?

No. Qovery abstracts all Kubernetes complexity. You interact with applications, databases, and environment variables - not pods, deployments, or ingress controllers. The AI agent handles the Kubernetes-specific configuration automatically.

Which cloud providers are supported?

AWS (EKS), Google Cloud (GKE), Azure (AKS), and Scaleway (Kapsule). You can also bring your own Kubernetes cluster (BYOK).

What about data migration?

Database schema and application code are migrated automatically. For data transfer (copying production data), you’ll need to handle the database dump/restore process. Qovery databases support standard import tools (pg_restore, mysql import, mongorestore).

How to create a custom domain?

Check the custom domain configuration documentation for setup instructions.

How to monitor my apps?

Qovery provides built-in observability including logs, metrics, and alerts. You can also integrate with Datadog, New Relic, or other monitoring tools.

Do you have preview environments?

Yes. Qovery provides Preview Environments that automatically create a clone of your environment on every pull request.

How to rollback?

App rollback is available through the deployment actions. You can roll back to any previous deployment in one click.

How does auto-scaling work?

Auto-scaling is configured in the application settings. Qovery supports horizontal pod autoscaling based on CPU/memory usage.

Can I use Terraform?

Yes. Qovery has a Terraform provider for infrastructure-as-code deployment. The AI agent can also deploy via Terraform if you choose that option during setup.

Is it possible to get a shell into my app?

Yes, via the cloud shell button in the Qovery Console or through the CLI with qovery shell.

How to connect to external services (MongoDB Atlas, AWS services)?

Use VPC peering for secure, private connectivity to external services in your cloud account.

Next Steps

Deploy with AI Agent

Full guide on installing and using Qovery AI skills

Production Setup

Set up a production-ready environment with best practices

Preview Environments

Automatically create environments for every pull request

Book a Demo

Get cloud architect assistance for your migration

Get Started

Platform Capabilities

How-to Guides

Tutorials

Troubleshooting

Useful Resources

Migrate to Kubernetes

Overview

Why Migrate with Qovery?

AI-Powered Migration

No Kubernetes Expertise

Your Cloud, Your Control

Cloud Architect Assistance

Compliance Ready

Cost Savings

Migrate from Any Platform

Heroku

Render

Railway

OpenShift

VMs / Bare Metal

Docker Compose

Other PaaS

How It Works

Getting Started

Option 1: AI-Powered Migration (Recommended)

Option 2: Manual Migration

Need Help? Talk to a Cloud Architect

Book a Migration Demo

What Gets Migrated?

FAQ

Next Steps

Deploy with AI Agent

Production Setup

Preview Environments

Book a Demo

​Overview

​Why Migrate with Qovery?

AI-Powered Migration

No Kubernetes Expertise

Your Cloud, Your Control

Cloud Architect Assistance

Compliance Ready

Cost Savings

​Migrate from Any Platform

Heroku

Render

Railway

OpenShift

VMs / Bare Metal

Docker Compose

Other PaaS

​How It Works

​Getting Started

​Option 1: AI-Powered Migration (Recommended)

​Option 2: Manual Migration

​Need Help? Talk to a Cloud Architect

Book a Migration Demo

​What Gets Migrated?

​FAQ

​Next Steps

Deploy with AI Agent

Production Setup

Preview Environments

Book a Demo

Overview

Why Migrate with Qovery?

Migrate from Any Platform

How It Works

Getting Started

Option 1: AI-Powered Migration (Recommended)

Option 2: Manual Migration

Need Help? Talk to a Cloud Architect

What Gets Migrated?

FAQ

Next Steps