This tutorial covers the OpenStack installation from the Ubuntu 12.10 Server Edition CD, and assumes a basic network topology, with a single system serving as the "all-in-one cloud infrastructure".Due to the tutorial's simplicity, the instructions as-is are not intended to set up production servers although it allows you to have a POC (proof of concept) of the Ubuntu Cloud using OpenStack.

To deploy a minimal Ubuntu Cloud infrastructure, you'll need at least:

The "Minimum Topology" recommended for production use is using three nodes - One master server running nova services (except compute) and two servers running nova-compute. This setup is not redundant and the master server is a SPoF (Single Point of Failure).

Before we start installing OpenStack we need to make sure we have bridging support installed, a MySQL database, and a central time server (ntp). This will assure that we have instantiated machines and hosts in sync.

In this example the "private network" will be in the 10.0.0.0/24 range on eth1. All the internal communication between instances will happen there while the "public network" will be in the 10.153.107.0/29 range on eth0.

OpenStack Compute (Nova) is a cloud computing fabric controller (the main part of an IaaS system). It is written in Python, using the Eventlet and Twisted frameworks, and relies on the standard AMQP messaging protocol, and SQLAlchemy for data store access.

Install OpenStack Nova components

sudo apt-get install nova-api nova-network nova-volume nova-objectstore nova-scheduler \
nova-compute euca2ools unzip

Restart libvirt-bin just to make sure libvirtd is aware of ebtables.

sudo service libvirt-bin restart

Install RabbitMQ - Advanced Message Queuing Protocol (AMQP)

sudo apt-get install rabbitmq-server

Edit /etc/nova/nova.conf and add the following:

# Nova config FlatDHCPManager
--sql_connection=mysql://novauser:novapassword@localhost/nova
--flat_injected=true
--network_manager=nova.network.manager.FlatDHCPManager
--fixed_range=10.0.0.0/24
--floating_range=10.153.107.72/29
--flat_network_dhcp_start=10.0.0.2
--flat_network_bridge=br100
--flat_interface=eth1
--public_interface=eth0

Restart OpenStack services

for i in nova-api nova-network nova-objectstore nova-scheduler nova-volume nova-compute; \
do sudo stop $i; sleep 2; done

for i in nova-api nova-network nova-objectstore nova-scheduler nova-volume nova-compute; \
do sudo start $i; sleep 2; done

Migrate Nova database from sqlite db to MySQL db. It may take a while.

sudo nova-manage db sync

Define a specific private network where all your Instances will run. This will be used in the network of fixed Ips set inside nova.conf .

sudo nova-manage network create --fixed_range_v4 10.0.0.0/24 --label private \
--bridge_interface br100

Define a specific public network and allocate 6 (usable) Floating Public IP addresses for use with the instances starting from 10.153.107.72.

sudo nova-manage floating create --ip_range=10.153.107.72/29

Create a user (user1), a project (project1), download credentials and source its configuration file.

cd ; mkdir nova ; cd nova
sudo nova-manage user admin user1
sudo nova-manage project create project1 user1
sudo nova-manage project zipfile project1 user1
unzip nova.zip
source novarc

Verify the OpenStack Compute installation by typing:

sudo nova-manage service list
sudo nova-manage version list

If nova services don't show up correctly restart OpenStack services as described previously. For more information please refer to the troubleshooting section on this guide.

Nova uses Glance service to manage Operating System images that it needs for bringing up instances. Glance can use several types of storage backends such as filestore, s3 etc. Glance has two components - glance-api and glance-registry. These can be controlled using the concerned upstart service jobs. For this specific case we will be using mysql as a storage backend.

Install Glance

sudo apt-get install glance

Create a database and user for glance

sudo mysql -uroot -ppassword -e "CREATE DATABASE glance;"
sudo mysql -uroot -ppassword -e "GRANT ALL ON glance.* TO glanceuser@localhost \
IDENTIFIED BY 'glancepassword';"

Edit the file /etc/glance/glance-registry.conf and edit the line which contains the option "sql_connection =" to this:

sql_connection = mysql://glanceuser:glancepassword@localhost/glance

Remove the sqlite database

rm -rf /var/lib/glance/glance.sqlite

Restart glance-registry after making changes to /etc/glance/glance-registry.conf. The MySQL database will be automatically populated.

sudo restart glance-registry

If you find issues take a look at the log file in /var/log/glance/api.log and /var/log/glance/registry.log.

Before you can instantiate images, you first need to setup user credentials. Once this first step is achieved you also need to upload images that you want to run in the cloud. Once you have these images uploaded to the cloud you will be able to run and connect to them. Here are the steps you should follow to get OpenStack Nova running instances:

Download, register and publish an Ubuntu cloud image

wget http://cloud-images.ubuntu.com/raring/current/raring-server-cloudimg-amd64.tar.gz
cloud-publish-tarball raring-server-cloudimg-amd64.tar.gz raring_amd64

Create a key pair and start an instance

cd ~/nova
source novarc
euca-add-keypair user1 > user1.priv
chmod 0600 user1.priv

Allow icmp (ping) and ssh access to instances

euca-authorize default -P tcp -p 22 -s 0.0.0.0/0
euca-authorize -P icmp -t -1:-1 default

Run an instance

ami=`euca-describe-images |  awk {'print $2'} | grep -m1 ami`
euca-run-instances $ami -k user1 -t m1.tiny
euca-describe-instances

Assign public address to the instance.

euca-allocate-address
euca-associate-address -i instance_id public_ip_address
euca-describe-instances

You must enter above the instance_id (ami) and public_ip_address shown above by euca-describe-instances and euca-allocate-address commands.

Now you should be able to SSH to the instance

ssh -i user1.priv ubuntu@ipaddress

To terminate instances

euca-terminate-instances instance_id

Swift is a highly available, distributed, eventually consistent object/blob store. It is used by the OpenStack Infrastructure to provide S3 like cloud storage services. It is also S3 api compatible with amazon.

Organizations use Swift to store lots of data efficiently, safely, and cheaply where applications use an special api to interface between the applications and objects stored in Swift.

Although you can install Swift on a single server, a multiple-server installation is required for production environments. If you want to install OpenStack Object Storage (Swift) on a single node for development or testing purposes, use the Swift All In One instructions on Ubuntu.

For more information see: http://swift.openstack.org/development_saio.html .

Community Support

The Ubuntu Cloud documentation uses terminology that might be unfamiliar to some readers. This page is intended to provide a glossary of such terms and acronyms.