Skip to content

5 ClusterConfiguration

GitHub-Actions edited this page Aug 5, 2020 · 3 revisions

Home

This step requires steps :

Generating a fixed IP address

In order to expose the SFTP server, you need to reserve a static IP address that will be used by Kubernetes. The IP address needs to be in the same zone as the Kubernetes cluster.

For Google Cloud :

gcloud compute addresses create ${APP_NAME}-ip --region ${K8S_REGION} --ip-version IPV4

Once this is done, you can update the APP_SERVICE_IP environment variable in the ./env/${ENV} file.

Deploying to the Kubernetes cluster

If you're using Kubernetes, the credentials should be provided through mounted secrets volume.

This project contains a Helm Chart that can deploy the service to a Kubernetes cluster.

First connect kubectl to the Kubernetes cluster by using the following command :

make setup_kubernetes_access

You can setup Helm on your cluster, if not already done, by running:

make helm_setup

Once that's done, you can just run:

make helm_install

This will automatically pick up any changes made to the relevant files in ./config/${ENV} and ./credentials/${ENV}, regenerate config, generate the approriate values.yaml file for Helm, and deploy the changes to your Kubernetes cluster.