Itnet Technologies
Expertise
About
Book a meeting
ITNET
ITNET Technologies
Online
Nola

Welcome!

Before we start, introduce yourself so Nola can better assist you.

France

Your data remains confidential

ITNET TECHNOLOGIES

Sovereign cloud - cybersecurity - datacenter

A technical partner for your critical digital environments.

ITNET TECHNOLOGIES designs, hosts and secures cloud, cybersecurity and datacenter infrastructure for organizations that require sovereignty, availability and operational control.

Plan an IT auditExplore sovereign cloud

Business contact

Emailcontact@itnet-technologies.comPhone+33 9 86 55 06 55
Head office22 Rue de Pissefontaine, 78570 Chanteloup-les-Vignes
Dubai DIFC officeDubai International Financial Centre (DIFC), Dubai, United Arab Emirates
AvailabilityMon.-Fri. 09:00-18:00

Solutions

  • Sovereign cloud & secure hosting
  • Managed cybersecurity & audit
  • Immersion cooling
  • Direct Liquid Cooling
  • VOLTANEUM dielectric liquid
  • AXMARIL secret management

Trust

  • French company, data hosted in France depending on project scope
  • Architectures aligned with GDPR, NIS2, ISO 27001 and HDS requirements to scope
  • Monitoring and support for critical services
  • Infrastructure designed for performance and energy efficiency

Company

  • Book a meeting
  • Invest in ITNET
  • Resources & news

Legal

  • Legal notice
  • Privacy policy

Follow ITNET

LinkedInYouTubeX
SASU - SIRET 890 177 470 00014
Cloud, cybersecurity and sustainable infrastructure

Certifications, frameworks and technical assurances

Trust markers for your critical infrastructure.

Certifications & tools

Datacenter, security & compliance

© 2026 ITNET TECHNOLOGIES. All rights reserved.

Designed and operated by ITNET TECHNOLOGIES.

Back to BlogBlog

Deploying DevStack: a practical guide to building an OpenStack test environment

Learn how to deploy DevStack on Ubuntu to quickly build an OpenStack development, test, and validation environment.

Mouhamed BANKOLEIT Infrastructure Expert
March 18, 20263 min read
Deploying DevStack: a practical guide to building an OpenStack test environment

Deploying DevStack: a practical guide to building an OpenStack test environment

If you want to discover OpenStack, validate a private cloud architecture, or quickly test components without launching a full production platform, DevStack remains one of the fastest paths. It allows teams to deploy an OpenStack development environment in a few steps, with a workflow designed for testing, demos, and technical validation.

Why use DevStack

DevStack is especially useful to:

  • discover OpenStack,
  • test Nova, Neutron, Horizon, or Cinder,
  • validate integrations,
  • build a fast cloud lab,
  • reproduce technical scenarios without heavy industrial setup.

A key point must remain clear: DevStack is not built for production. It is first and foremost a lab tool.

Prerequisites before installation

To deploy DevStack cleanly, it is best to start from a dedicated Ubuntu VM or server with:

  • a clean and updated system,
  • sudo access,
  • at least 8 GB of RAM for a comfortable lab,
  • several vCPUs,
  • enough disk space,
  • a dedicated non-root user.

A noisy or overloaded environment quickly makes installation harder.

Deployment steps

1. Update the system

bash
sudo apt update && sudo apt upgrade -y
sudo apt install -y git

2. Create a dedicated user

bash
sudo useradd -s /bin/bash -d /opt/stack -m stack
echo "stack ALL=(ALL) NOPASSWD: ALL" | sudo tee /etc/sudoers.d/stack
sudo su - stack
Tags:#sdn

Share this article

Related articles

3. Clone DevStack

bash
git clone https://opendev.org/openstack/devstack
cd devstack

4. Prepare local.conf

ini
[[local|localrc]]
ADMIN_PASSWORD=ChangeMe123!
DATABASE_PASSWORD=$ADMIN_PASSWORD
RABBIT_PASSWORD=$ADMIN_PASSWORD
SERVICE_PASSWORD=$ADMIN_PASSWORD
HOST_IP=192.168.1.50

5. Launch the installation

bash
./stack.sh

At the end, you typically get a Horizon dashboard, admin access, and a functional OpenStack base to test instances, networks, and images.

Limits to keep in mind

DevStack is excellent for learning and testing, but it does not replace an industrialized OpenStack architecture. Its main limits are:

  • lower stability than production,
  • poor fit for critical workloads,
  • strong dependency on network and resource quality,
  • simplified maintenance but no long-term operating model.

Best practices

To get real value from DevStack:

  • isolate the machine,
  • version your local.conf,
  • start with a minimal scope,
  • document your tests,
  • never mix lab and production thinking.

Conclusion

Deploying DevStack is an excellent entry point for learning OpenStack, validating a private cloud scenario, or accelerating a testing phase. Used correctly, it helps teams build expertise, explore the main services, and prepare for a more robust architecture later on.

Main illustration

DevStack OpenStack visual
DevStack OpenStack visual

Illustration generated for this article and stored in Nextcloud.

FAQ

Is DevStack meant for production?

No. DevStack is designed for development, testing, demos, and technical validation.

Which OS is best for DevStack?

Ubuntu is one of the easiest choices to get started quickly.

Can you learn OpenStack with DevStack?

Yes. It is one of the best ways to understand OpenStack services and basic workflows.

What is the difference between DevStack and a real OpenStack deployment?

DevStack is a fast lab environment. A production deployment requires a durable, secure, and industrialized architecture.

📝
Blog
July 2, 20267 min

Voltaneum and private AI inference: placing GPU workloads at the right trust level

How to operate a sovereign GPU cloud by aligning AI placement, confidentiality, useful capacity and operating evidence.

Mouhamed BANKOLE
Read more
#voltaneum#cloud#datacenter
📝
Blog
July 2, 20266 min

Zero-trust VPS: reducing attack surface without blocking operations

A field-ready approach to secure exposed VPS services while preserving the speed expected from cloud delivery.

Mouhamed BANKOLE
Read more
#vps
📝
Blog
July 2, 20266 min

Immersion GPU inference: measuring useful capacity before promising performance

A practical frame to turn GPU density into a stable, measurable and operable AI service.

Mouhamed BANKOLE
Read more