Deploy multi node OpenStack on Virtual Machines.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Andreas Jaeger 2585b12b87 Cleanup py27 support 2 years ago
doc Cleanup py27 support 2 years ago
labs Update Ubuntu LTS ISO to 18.04.4 2 years ago
playbooks/scripts Use Zuul v3 native job for building scripts 4 years ago
tools Generate windows scripts using in the CI 4 years ago
.gitignore Add Apple OS X ".DS_Store" to ".gitignore" file 3 years ago
.gitreview OpenDev Migration Patch 3 years ago
.zuul.yaml Remove packaging 3 years ago
CONTRIBUTING.rst Refactors the architecture of osbash 6 years ago
HACKING.rst Sync Sphinx requirement 2 years ago
LICENSE Refactors the architecture of osbash 6 years ago
README.rst Replace URLs with URLs 2 years ago
bindep.txt Update bindep.txt for doc builds 4 years ago Refactors the architecture of osbash 6 years ago
tox.ini Cleanup py27 support 2 years ago


Team and repository tags


Training labs


Training-labs provides an automated way to deploy Vanilla OpenStack, closely following the OpenStack Install Guide.

Training-labs offers an easy way to set up an OpenStack cluster which is a good starting point for beginners to learn OpenStack, and for advanced users to test out new features, and check out different capabilities of OpenStack.

On top of that training-labs is also a good way to test the installation instructions on a regular basis.

Training-labs is a project under OpenStack Documentation. For more information see the OpenStack wiki.


VirtualBox is the default hypervisor used by training-labs. Alternatively, you can use KVM (just set PROVIDER=kvm in labs/config/localrc).

Getting the Code for an OpenStack Release

The current release is master which usually deploys the current stable OpenStack release. Unless you have a reason to go with an older release, we recommend using master.

For non-development purposes (training, etc.), the easiest way to get the code is through downloading the desired archive from OpenStack Training Labs. Unpack the archive and you are good to go.

How to run the scripts for GNU/Linux and macOS

Change directory:

$ cd training-labs/labs/

By default, the cluster is built on Virtualbox VMs.

Run the script by:

$ ./ -b cluster

How to run the scripts for Windows

The easiest and recommended way to get everything you need besides VirtualBox is to download a zip file for Windows from the Training Labs page.

The zip files include pre-generated Windows batch files.

Creates the host-only networks used by the node VMs to communicate:

> create_hostnet.bat

Creates the base disk:

> create_base.bat

Creates the node VMs based on the base disk:

> create_ubuntu_cluster_node.bat

What the script installs

Running this will automatically spin up 2 virtual machines in VirtualBox/KVM:

  • Controller node
  • Compute node

Now you have a multi-node deployment of OpenStack running with the following services installed.

  • Keystone
  • Nova
  • Neutron
  • Glance
  • Cinder
  • Horizon

How to access the services

There are two ways to access the services:

  • OpenStack Dashboard (horizon)

You can access the dashboard at:

Admin Login:

  • Username: admin
  • Password: admin_pass

Demo User Login:

  • Username: demo
  • Password: demo_pass

You can ssh to each of the nodes by:

# Controller node
$ ssh osbash@

# Compute node
$ ssh osbash@

Credentials for all nodes:

  • Username: osbash
  • Password: osbash

After you have ssh access, you need to source the OpenStack credentials in order to access the services.

Two credential files are present on each of the nodes:


Source the following credential files

For Admin user privileges:

$ source

For Demo user privileges:

$ source

Note: Instead 'source' you can use '.', or you define an alias. Now you can access the OpenStack services via CLI.


To review specifications, see Training-labs

Mailing lists, IRC

To contribute, join the IRC channel, #openstack-doc, on IRC freenode or write an e-mail to the OpenStack Development Mailing List Please use [training-labs] tag in the subject of the email message.

You may have to subscribe to the OpenStack Development Mailing List to have your mail accepted by the mailing list software.

Sub-team leads

Feel free to ping Roger, Julen, or Pranav via email or on the IRC channel #openstack-doc regarding any queries about training-labs.

  • Roger Luethi
    • Email:
    • IRC: rluethi
  • Pranav Salunke
    • Email:
    • IRC: dguitarbite
  • Julen Larrucea
    • Email:
    • IRC: julen, julenl


Training-labs uses the Doc Team Meeting:


Follow various links on training-labs here: