Podman and Kubernetes

Learn how to use Podman with Kubernetes — including generating Kubernetes YAML files, running them locally, and preparing workloads for production.

One of Podman’s most powerful features is its native support for Kubernetes YAML.

This makes it easy to develop container workloads locally and then export them directly to Kubernetes — no need to rewrite configs or convert tools.

📄 Generating Kubernetes YAML from Podman

You can create containers or pods with Podman and then export them as Kubernetes-compatible YAML.

podman generate kube mypod > mypod.yaml

This includes:

• Pod and container configuration
• Environment variables
• Ports and volumes
• Restart policies (as annotations)

🧠 The output is structured as a Pod definition — ready to use with kubectl or podman play kube.

▶️ Running Kubernetes YAML with Podman

Test Kubernetes manifests locally using:

podman play kube mypod.yaml

This will:

• Create a pod and containers from the YAML file
• Apply port mappings, environment variables, etc.
• Use Podman’s pod infrastructure

📦 Example Workflow

# Create a pod manually
podman pod create --name apppod -p 8080:80

# Add services
podman run -dt --pod apppod --name app nginx
podman run -dt --pod apppod --name db redis

# Export as Kubernetes YAML
podman generate kube apppod > apppod.yaml

# Recreate from YAML
podman play kube apppod.yaml

This lets you build your apps locally and deploy using the same manifests you’ll use in production.

🔄 Integration with Kubernetes Clusters

While Podman is not a Kubernetes runtime itself (like CRI-O), it plays a key role in:

• Prototyping and testing deployments
• Validating configurations before deploying to the cloud
• Generating boilerplate YAML for clusters

🧰 Podman works well alongside Kubernetes tools like kubectl, minikube, or kind.

🧪 Use Case: Local Dev + Cloud Deployment

1. Develop and test containers with Podman
2. Use generate kube to export configuration
3. Use kubectl apply to deploy to a real Kubernetes cluster

This simplifies testing and CI/CD pipelines, and bridges local workflows with production orchestration.

⚠️ Caveats

• podman play kube only supports basic Pod, Deployment, and Service objects
• More complex Kubernetes features (e.g., Ingress, ConfigMaps, secrets) require manual YAML editing or use of Kubernetes CLI

🧩 Podman + CRI-O

In Kubernetes environments, Podman’s sibling tool — CRI-O — serves as the container runtime. It uses the same container libraries and OCI standards.

🔗 If you’re comfortable with Podman, transitioning to CRI-O + Kubernetes will feel very familiar.

Next up: Podman in Production

< Previous Next >

You can use the arrows  ← → on your keyboard to navigate between lessons.