Kubernetes Manifest Generator
Fill in a short form and get Kubernetes YAML for a Deployment, Service and Ingress, with resources, health probes and environment variables.
How to generate Kubernetes YAML
Fill in the app
Name, image, replicas, port and health check path.
Choose how to expose it
A Service type, and an Ingress with a host name and HTTPS if you need one.
Apply it
Download app.yaml and run kubectl apply -f app.yaml.
Worked example
This is what Kubernetes Manifest Generator gives for its own sample values. Open the tool above and change anything to try your own.
- App name
- web
- Container image
- ghcr.io/acme/web:1.0.0
- Namespace (optional)
- Replicas
- 2
- Container port (0 for none)
- 8080
- Health check path
- /healthz
Manifests
apiVersion: apps/v1
kind: Deployment
metadata:
name: web
labels: &a1
app.kubernetes.io/name: web
spec:
replicas: 2
selector:
matchLabels: *a1
template:
metadata:
labels: *a1
spec:
...When to use Kubernetes Manifest Generator
You are deploying an app to Kubernetes for the first time and need a correct Deployment, Service and Ingress to start from.
A small service needs the same standard manifests as the rest, without copying and editing an old file.
Tips for a better Kubernetes Manifest Generator result
Set the health check path to an endpoint that answers quickly without touching the database, so probes do not fail under load.
Use a namespace per environment, and change only the image tag between them when you promote a release.
Kubernetes Manifest Generator: limits and what to check
The Ingress assumes an ingress controller is installed; without one, the Ingress object is created but nothing routes traffic.
HTTPS on the Ingress needs a TLS secret with that name, usually made by cert-manager, before browsers will connect.
About Kubernetes Manifest Generator
Get correct, current Kubernetes YAML for a typical web app without copying from old blog posts. The Deployment's selector matches its pod labels, the Service and Ingress point at the named port, resources are set, and readiness and liveness probes use your health path.
FAQs
Which API versions are used?
apps/v1 for the Deployment, v1 for the Service and ConfigMap, and networking.k8s.io/v1 for the Ingress, which current clusters support.
Why use a named port?
The Service and Ingress refer to the port by name, so changing the number later means editing one place.
Should I set CPU limits?
Many teams set memory limits but leave CPU limits off to avoid throttling. Requests should always be set.