Microk8s for Kubeflow

This document will outline steps that will get your local installation of Kubeflow running on top of Microk8s inside of a native Hypervisor. Multipass is used to create a the native VM. Microk8s is used to provide a simple, single-node Kubernetes cluster.

By the end of this document, you’ll have a local installation of a Kubernetes cluster along with all the default core components of Kubeflow deployed as services in the pods. You should be able to access JupyterHub notebooks, and the Kubeflow Dashboard.

Install Multipass

If you do not already have a multipass installed, install a new one.

Mac OS X

Install Multipass using the native Mac OS installer.

Linux

It can be installed with the following command on any snap-enabled linux:

$ sudo snap install multipass --beta --classic

Start an Ubuntu Virtual Machine

Download cloud-init file

This cloud-init file can be used repeatedly, each time a new multipass VM is created

wget https://bit.ly/2tOfMUA -O kubeflow.init
Start your Multipass VM

You can mount a local volume into the VM. The second command adds the local directory.

$ multipass launch bionic -n kubeflow -m 8G -d 40G -c 4 --cloud-init kubeflow.init
$ multipass mount . kubeflow:/multipass

Note: These are the minimum recommended settings on the VM created by Multipass for the Kubeflow deployment. You are free to adjust them higher based on your host machine capabilities and workload requirements.

Install Kubernetes

Log into the VM and install some basic supporting tools. This will install kubernetes, powered by microk8s, and other tools necessary to deploy Kubeflow

multipass shell kubeflow                      # log into vm
sudo /kubeflow/install-kubeflow-pre-micro.sh  # install microk8s, etc.

Install Kubeflow

This step assumes you’ve stored your github token in a file in the host machine. If you haven’t already done this, please put: export GITHUB_TOKEN=<your token> into /multipass/github-token.txt

source /multipass/github-token.txt   # exports your github token
/kubeflow/install-kubeflow.sh        # waits until all pods are “running”; prints the port

This script will print out the port number of JupyterHub.

Access Kubeflow

You can get the IP address of the VM using multipass list. Assuming you’ve used the same name (kubeflow) above, you can now access the JupyterHub page from your browser at http://:

For JupyterHub, you’ll be landing on a login page.

  • Use any username and password to login
  • Pick an available CPU tensorflow image
  • Provide at least 2 CPUs
  • Provide 4Gi for the memory
  • Leave “Extra Resource Limits” alone for now
  • Click Spawn.
  • You should be redirected to a page that waits while the server is starting.

If the page doesn’t refresh, please see troubleshooting.

Where to go next

Refer to the user guide