benpayflic/kube-cluster-do-creator

Provision a Kubernetes cluster on Digital ocean

1. Downlaod and install python 3 on local machine (if not already installed): https://www.python.org/downloads/

2. cd to project directory and create a virtual enviroment

   $ python3 -m venv env
   

3. Activate virtual enviroment

   $ source env/bin/activate
   

4. Install dependencies

   $ pip3 install -r requirements.txt
   

5. Create and add an SSH key to your digital ocean account: https://docs.digitalocean.com/products/droplets/how-to/add-ssh-keys/to-team/

6. Create a digital ocean access token for API requests: https://docs.digitalocean.com/reference/api/create-personal-access-token/

7. Add the token to your environment:

   $ export DIGITAL_OCEAN_ACCESS_TOKEN=[your_token]

8. Spin up the cluster

   $ bash create.sh

9. Add kube config to local machine Copy and paste the output from script below line "++++++ Copy the below output and paste into kube config on local machine ++++++" into the open text file. Save and close.

   $ cd ~/.kube
   

   $ open -a TextEdit config
   

10. Test local machine has cluster access

    $ kubectl get nodes -o wide
    

11. Create app namespace and set context

    $ kubectl create namespace {namespace} (example: elasticsearch-ns)
    

    $ kubectl config set-context kubernetes-admin@kubernetes --namespace={namespace name}
    

12. Install Traefik

    $ helm repo add traefik https://helm.traefik.io/traefik
    

    $ helm repo update
    

    $ helm install traefik traefik/traefik --namespace traefik --create-namespace
    

13. Access traefik dashboard

    $ kubectl get pods -n traefik
    

    $ kubectl port-forward {pod name} 9000:9000 -n traefik
    

    Go to http://127.0.0.1:9000/dashboard/#/ to ensure traefik is up and running.

14. Add all required K8 resorces to cluster using the command kubectl apply -f {path/to/yaml} Enjoy :)