OpenClaw on Ubuntu 24.04 LTS: Installation Requirements and Overview

OpenClaw is an open-source, self-hosted AI agent framework. Rather than waiting for a prompt like a chatbot, it runs continuously on a server, connects to messaging apps such as WhatsApp, Telegram, Discord, or Slack, and can carry out real tasks — reading and writing files, running shell commands, browsing the web, and managing email or a calendar — using whichever AI model provider you connect it to (Anthropic Claude, OpenAI, Google Gemini, or a local model through Ollama).

Installation is streamlined through a 1-click process. This guide covers the underlying requirements and manual installation steps, which are useful for documentation, troubleshooting, and technical support.

Prerequisites

Before installing OpenClaw on Ubuntu 24.04 LTS, make sure your system meets the following requirements:

Operating System: Ubuntu 24.04 LTS (64-bit)
Access Level: Root or sudo privileges
Network Access: Outbound internet connectivity for downloading packages and reaching your chosen AI model provider
Runtime: Node.js is required to run OpenClaw. The installer script installs a compatible version automatically, so a separate Node.js setup step is only needed for a manual or from-source install.

OpenClaw does not publish a fixed minimum CPU/RAM specification. A modest VPS is sufficient when OpenClaw is connected to a cloud AI provider (Claude, GPT, Gemini) doing the reasoning. Running a local model on the same server through Ollama requires significantly more RAM and CPU, scaling with the size of the local model.

Required Packages and Dependencies

To install OpenClaw on Ubuntu 24.04 LTS, ensure the following are available:

curl – to download the installer
git – only if installing from the source repository instead of the installer script
Node.js and npm – OpenClaw's runtime; the installer script installs this for you

Docker and a web server (Nginx/Apache) are not required for a standard OpenClaw install. OpenClaw's control dashboard is served locally by the OpenClaw gateway process itself, not by a separate web server. Docker is only relevant if you deliberately choose a container-based deployment method instead of the installer script.

Installation Steps (Manual Reference)

These steps are for documentation and technical support. End users typically use a 1-click installation process handled by the internal team.

Update your system:

sudo apt update && sudo apt upgrade -y
Install curl and git:

sudo apt install -y curl git
Install OpenClaw
1. Using the installer script (recommended — installs Node.js and OpenClaw together):
  
  curl -fsSL https://openclaw.ai/install.sh | bash
2. Using npm, if Node.js is already installed on the server:
  
  npm i -g openclaw
3. From source, for troubleshooting or contribution purposes:
  
  git clone https://github.com/openclaw/openclaw.git
  cd openclaw
  corepack enable
  pnpm install
  pnpm openclaw onboard
Run the guided setup

openclaw onboard

This interactive wizard configures the model provider (API key or OAuth for Claude, GPT, or Gemini), the workspace, and the gateway (the local control dashboard, bound to port 18789 by default). It can also offer to install OpenClaw as a background service (systemd on Linux) so it keeps running after you disconnect.
Access the dashboard on a headless VPS
Since a Self-Managed VPS has no graphical desktop, OpenClaw runs headless and binds its control dashboard to 127.0.0.1:18789 — it is not reachable directly over the internet. To view it from your own computer, open an SSH tunnel:

ssh -L 18789:127.0.0.1:18789 your-username@your-server-ip

Then open http://127.0.0.1:18789 in a browser on your local machine.
Verify Installation

openclaw --version
openclaw doctor
openclaw gateway status

Example Output:
```
$ openclaw --version
openclaw 4.2.0

$ openclaw doctor
✔ Node.js version OK
✔ Configuration file valid
✔ No issues found

$ openclaw gateway status
Gateway: running (pid 18422)
Listening: 127.0.0.1:18789
```
openclaw doctor checks for configuration issues, and openclaw gateway status confirms the gateway process is running and reachable.

1-Click Install Process (Internal Team)

The internal team automates the installation of OpenClaw on Ubuntu 24.04 LTS, including:

Installing Node.js and OpenClaw via the installer script
Completing the onboarding/model-provider configuration
Starting the gateway as a background service

This ensures a seamless experience for end users without manual setup or SSH tunneling.

Firewall Notes

Because the OpenClaw dashboard binds to 127.0.0.1 by default, no inbound firewall rule is required to protect it — it simply isn't exposed externally. If a messaging channel integration requires an inbound webhook, only open the specific port that integration documents, and close it again if the integration is removed.

Troubleshooting and Support

If you encounter issues during installation or setup:

Run openclaw doctor to check for configuration problems
Run openclaw gateway status to confirm the gateway process is running and reachable
Confirm the server has outbound network connectivity to your chosen AI model provider
Confirm the SSH tunnel is active if the dashboard isn't loading locally
Ensure proper user permissions for the account running OpenClaw

Summary

Installing OpenClaw on Ubuntu 24.04 LTS means installing Node.js and the OpenClaw CLI, then running the guided onboarding wizard to connect a model provider and start the gateway. While most users rely on the automated 1-click setup, understanding the manual process — and the fact that the dashboard is local-only by design — helps with troubleshooting and technical support.

Did you find this article helpful?