The Nectar Container Orchestration Engine (COE) service provides support for provisioning Kubernetes clusters as a service on the Nectar Research Cloud using OpenStack Magnum. This allows a user to spin up a container cluster (Kubernetes) on Nectar Research Cloud.

Currently only Kubernetes as a Container Orchestration Engine is supported. For a tutorial on how to create a Kubernetes cluster, please visit the Kubernetes Tutorial.

To spin up a Kubernetes container cluster, you need quotas for the following resources:    

  • Clusters - 1 per cluster that you want to run
  • Floating IPs - 2 per cluster
  • Load balancers - 3 per cluster
  • Networks - 1 per cluster
  • Instances - at least 2 per cluster

There is a step by step tutorial available to teach you how to create a Cluster. Please visit the Nectar Tutorial Website.  

Creating a Cluster

You can create a cluster using either the Dashboard or CLI tools.

Using Dashboard 

  1. Log on to the Nectar Dashboard
  2. Navigate to Container Infra.
  3. Click on Clusters, then Create Cluster.  This will open the Create dialog
  4.  On the Info tab, give your cluster a name and choose a cluster template.
    We have pre-defined global templates (in format kubernetes-{az}-{version}) to help you get started.
    Choose the template that you want your cluster to be in.
  5.  On the Misc tab, make sure you select your Keypair.
  6.  Click Submit.

Using CLI

  1. Install python-magnumclient. You need python-magnumclient >= 2.9.0

    pip install python-magnumclient

  2.  Boot a cluster    

    openstack coe cluster create --cluster-template kubernetes-melbourne-v14.6 \
    --keypair <mykey> mycluster

Operating your Cluster

Once your cluster is up (NOTE: It takes about 20 mins for a cluster to build),you can control it using kubectl.

  1.   Install kubectl
  2.   Set up the credentials to connect to the cluster. Firstly, create an configdir

    mkdir ~/kubernetes/
    cd ~/kubernetes/

  3.   Create the config files    

    openstack coe cluster config mycluster

  4. Set the ENV by copying the output from the previous command  

    export KUBECONFIG=$HOME/kubernetes/config

  5.   Use kubectl to connect to it

   kubectl get all

Operating your Cluster using web interface

Alternatively, you can also administer it from the web.

  1. Set up a role for the service account
    kubectl create clusterrolebinding kubernetes-dashboard --clusterrole=cluster-admin --serviceaccount=kube-system:kubernetes-dashboard
  2. List the secrets
    kubectl -n kube-system get secret
  3. Get the secret token. It will be in format kubernetes-dashboard-token-XXXXX
    kubectl -n kube-system describe secret kubernetes-dashboard-token-XXXXX
  4. Copy the token
  5. Start the web interface
    kubectl proxy
  6. In your browser, go to the following URL
  7. Use your token to log in



Availability Zone

You can boot in a different availability zone by using --labels. E.g.

openstack coe cluster create --cluster-template mytemplate \
--keypair <mykey> --labels availability_zone <AZ> mycluster