Skip to content

Latest commit

 

History

History
32 lines (25 loc) · 2.42 KB

README.md

File metadata and controls

32 lines (25 loc) · 2.42 KB
Raw

The deploy workflow is used to deploy the application to kubernetes cluster.

Setup kubectl with github action

The workflow uses https://github.com/Azure/k8s-set-context for installing kubectl and also configure the context.

The action requires the kubeconfig property set with the string value of k8s cluster config. So simply create the secret with name KUBECONFIG and copy the value of the kube config from ~/.kube/config to the secret value.

Note: Since the ~/.kube/config may have multiple k8s cluster configurations. So its always better to download the k8s cluster specific config from cloud provider and use it's value as the value of KUBECONFIG secret variable.

Setting secrets for deploy job

Since the deploy job makes use of environments. We can create the secrets inside the environment settings. Following steps will help you setup the secrets required for deployment. The workflow makes use of https://github.com/Azure/k8s-create-secret action.

  • Go to Repo Settings > Environments > Production
  • Creating secrets of type generic or kubernetes.io/dockerconfigjson is easy. If you already have the secret in k8s then just run this command - kubectl get secret <SECRET_NAME> -o jsonpath='{.data}' -n <NAMESPACE> then copy the output to the secret value inside that specific environment secret on github. Then provide this secret to k8s-create-secret action as a value of data attribute.
  • To get the kubernetes config of the current context, run below command 
    kubectl config view --minify --flatten
    Then copy the output to KUBECONFIG secret.

Make sure you have setup below secrets inside Production environment configurations

  • AUTH_SECRET
  • KUBECONFIG
  • POSTGRES_SECRET
  • REGCRED_SECRET

NOTE: For kubernetes.io/dockerconfigjson type secret make sure you specify the type attribute as kubernetes.io/dockerconfigjson and not generic

Using a new k8s cluster for deployment

  • Make sure that you have updated the value of KUBECONFIG environment secret with the kubeconfig of newly created cluster.
  • If you own a digital ocean cluster then you can simply download the kube config file cluster overview. The contents of this file can be copied directly to KUBECONFIG github secret
  • After the first deployment on new cluster, update the domain's A record to point to the newly created load balancer.