Skip to content

Latest commit

 

History

History
150 lines (121 loc) · 4.75 KB

README.md

File metadata and controls

150 lines (121 loc) · 4.75 KB
Raw

Kubernetes Deployment Files

The deployment directory contains all files needed to deploy the ODS to a Kubernetes cluster.

Current Features

  • Generate deployment files for different namespaces and urls.
  • Deploy a new ODS instance to Kubernetes
  • Undeploy ODS instances

Planned Features

  • Rolling updates
  • Persistent volumes for databases
  • Scaling

Getting Started

In order to deploy to Kubernetes, kubectl must be installed and configured with a Kubernetes cluster.

  1. Use the generate-kubernetes-files.py script to create Kubernetes files from given templates.
    The script should be used as follows:

      usage: generate-kubernetes-files.py [-h] [--output_dir OUTPUT_DIR]
                                      [--template_dir TEMPLATE_DIR]
                                       namespace base_url

    This script reads the template files and replaces the variables for namespace and base url before saving the files in the output directory.

  2. Run the deploy.sh script.

    deploy.sh [-h] [-i input_dir] namespace -- Script to deploy Kubernetes files to a new namespace.

where:
    -h  show this help text
    -i  set the input directory for kubernetes files(default: ./)

    This script takes the previously generated Kubernetes files and deploys them.
    This contains the following steps:

    • Recreate the namespace to create a new clean environment
    • Copy the registry secrets from the ods namespace to the new namespace.
    • Deploy to Kubernetes

Deployment Evironment (Kubernetes) Setup

Micro-K8s Setup on Server

Do the following steps on a fresh Ubuntu Server installation:

  1. Run the iptables-config.sh script, in

  2. Check if the firewall behaviour is correct and persist settings with:

    echo "iptables-save > /etc/iptables/rules.v4"
echo "ip6tables-save > /etc/iptables/rules.v6"

  3. Install latest microk8s

    snap install microk8s --classic

  4. [Optional] Alias microk8s.kubectl to kubectl (This saves some writing and makes following Kubernetes examples easier).

    snap alias microk8s.kubectl kubectl

  5. Enable DNS, Dashboard and Ingress Addons

    microk8s.enable dns dashboard ingress

  6. Check if everything worked with

    microk8s.status

Create service accounts and access the cluster via remote connection

This tutorial is based on this one.

  1. Create Service Accounts

    kubectl apply -f - <<EOF
apiVersion: v1
kind: ServiceAccount
metadata:
  name: <replace with your desired account name>
EOF

  2. Check if service account was created

    kubectl get serviceaccounts/<replace with your account name> -o yaml

  3. Fetch the name of your secret

    kubectl describe serviceAccounts

    Note down the Mountable secrets information of your account, which has the name of the secret that holds the token

  4. Get the token for your service account

    kubectl describe secrets <replace with your secret>

    Note down the TOKEN value

  5. In the next step you have to retrieve the cluster information. To do so, execute the following:

    kubectl config view --flatten --minify > cluster-cert.txt  
cat cluster-cert.txt

    Note down the server and certificate-authority-data values.

For the next steps you need to have kubectl installed on your machine.

  1. Create a $HOME/.kube/config file on your machine with the following content:

    apiVersion: v1  
kind: Config  
users:  
- name: <replace this with your user name>  
	user:  
		token: <replace  this  with token  info>  
clusters:  
- cluster:
	certificate-authority-data: <replace this with certificate-authority-data info>  
	server: <replace this with server info>  
  name: ods-deployment-cluster
contexts:  
- context:  
   cluster: ods-deployment-cluster
   user: <replace with your user name>  
 name: ods-deployment-context  
current-context: ods-deployment-context

  2. Save and load the new context with the following command:

    kubectl config --kubeconfig=$HOME/.kube/config set-context ods-deployment-context

  3. Test if everything worked by running

     kubectl --version

Connect to the dashboard

The recommended way to connect to the dashboard is to use a proxy to connect to the api server.

Run kubectl proxy --port=8080 to start the proxy. Now kubectl takes care authenticating the proxy at the api server, to allow a secure communication over https.

Now you can reach the dashboard from the machine that runs the proxy under this link.

You need to login to the dashboard with your service account data.