Deploy Amplication to a Kubernetes cluster using Helm

Follow this article to deploy an Amplication server to a Kubernetes cluster using Helm.

The article focuses on deployment on a local server for testing and development purposes, but the same method can be implemented for a production environment.

Prerequisites

Before you start you need to install the following.

1. Kubernetes Cluster (We will be using the Docker Desktop K8s implementation)
2. Helm
3. Kubectl

Below we will go through each install step individually.

Install Kubernetes cluster

There are a few ways to install a Kubernetes cluster locally, in our example we'll be using Docker's Kubernetes Plugin (Docker Desktop). However there are a few other Open Source alternatives (e.g. Kind and Minikube), if you prefer to install a different K8s cluster, you can reference the list here here.

After you've finished installing Docker Desktop you can now enable the Kubernetes Plugin by going to the Kubernetes tab in the Docker Desktop settings page and check Enable Kubernetes

Install Helm and Kubectl

You will also need Helm and Kubectl installed.

1. Kubectl
2. Helm

Installing Amplication Helm Chart

Open the terminal and Pull down Amplication then go to ./amplication/chart/amplication directory.

1. git clone https://github.com/amplication/amplication.git
2. cd amplication/chart/amplication

The amplication chart is shipped with a set of defaults that should get you running without any additional configuration. Run this command

helm install amplication . --namespace amplication --create-namespace

Amplication Helm chart includes:

1. bitnami/postgresql server
2. Migration job for Amplication server.
3. A Docker Registry and Docker builder using https://hub.docker.com/_/docker, and a volume for the registry (docker image storage).
4. The Amplication server and a volume for /artifacts
5. The amplication scheduler

You can monitor the whole list of resources by running the following:

$ kubectl get all --namespace amplication

# Example output when all resources are completed and ready

NAME READY STATUS RESTARTS AGE

pod/amplication-app-5cb4f9d576-ldsvz 1/1 Running 0 53s

pod/amplication-app-migrate-f6xp4 0/1 Completed 0 53s

pod/amplication-docker-dd5bbd98f-5mrtj 1/1 Running 0 53s

pod/amplication-registry-5f969c6554-c4cbs 1/1 Running 0 53s

pod/amplication-scheduler-7b9557cdc5-dmjv2 1/1 Running 0 53s

pod/localpostgres-0 1/1 Running 0 53s

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE

service/amplication-app ClusterIP xx.xx.xx.xx <none> 3000/TCP 53s

service/amplication-docker ClusterIP xx.xx.xx.xx <none> 2375/TCP 53s

service/amplication-registry ClusterIP xx.xx.xx.xx <none> 5000/TCP 53s

service/localpostgres ClusterIP xx.xx.xx.xx <none> 5432/TCP 53s

service/localpostgres-headless ClusterIP None <none> 5432/TCP 53s

NAME READY UP-TO-DATE AVAILABLE AGE

deployment.apps/amplication-app 1/1 1 1 53s

deployment.apps/amplication-docker 1/1 1 1 53s

deployment.apps/amplication-registry 1/1 1 1 53s

deployment.apps/amplication-scheduler 1/1 1 1 53s

NAME DESIRED CURRENT READY AGE

replicaset.apps/amplication-app-5cb4f9d576 1 1 1 53s

replicaset.apps/amplication-docker-dd5bbd98f 1 1 1 53s

replicaset.apps/amplication-registry-5f969c6554 1 1 1 53s

replicaset.apps/amplication-scheduler-7b9557cdc5 1 1 1 53s

NAME READY AGE

statefulset.apps/localpostgres 1/1 53s

NAME COMPLETIONS DURATION AGE

job.batch/amplication-app-migrate 1/1 22s 53s

Port-Forwarding

The final step is port-forwarding the amplication-app port to your localhost so you can access it on a browser client.

kubectl port-forward deployment/amplication-app 3000:3000 --namespace amplication

Now you can access amplication locally on your browser at http://localhost:3000 👍