6.4 KiB
title
| title |
|---|
| Deploy InvenTree |
Deploying InvenTree
The development server provided by the Django ecosystem may be fine for a testing environment or small contained setups. However special consideration must be given when deploying InvenTree in a real-world environment.
Django apps provide multiple deployment methods - see the Django documentation.
There are also numerous online tutorials describing how to deploy a Django application either locally or on an online platform.
The following instructions provide a reasonably performant server, using gunicorn as a webserver, and supervisor as a process manager.
Install System Packages
Install required system packages (as superuser).
First, install required system packages as per the OS requirements.
Next, install the system packages required for your particular database:
MySQL
sudo apt-get install mysql-server libmysqlclient-dev
Postgresql
sudo apt-get install postgresql postgresql-contrib libpq-dev
Initial Setup
Create InvenTree User
It is highly recommended that the InvenTree server is not run under root. Create a user account from which we will run the server:
sudo useradd --create-home inventree
InvenTree source code, log files, etc will be located under the /home/inventree/ directory.
Switch to the inventree user so commands are performed in the correct context:
sudo su inventree
Download Source Code
Download InvenTree source code, into the ./src directory:
cd /home/inventree
git clone https://github.com/inventree/inventree src
Create Virtual Environment
Create a python virtual environment for installting required Python packages and binaries:
python3 -m venv env
source ./env/bin/activate
The shell prompt should now display the (env) prefix.
Create Required Directories
mkdir log static media backup
This step creates directories required by InvenTree:
- log - Store InvenTree log files
- static - Location of static files for the web server
- media - Location of uploaded media files
- backup - Location of database backup files
Install InvenTree Packages
The Python packages required by the InvenTree server must be installed into the virtual environment.
Run the invoke install command (from within the src directory):
cd src
invoke install
This installs all required Python packages using pip package manager. It also creates a (default) database configuration file which needs to be edited to meet user needs before proceeding (see next step below).
Additionally, this step creates a SECRET_KEY file which is used for the django authentication framework.
!!! warning "Keep it secret, keep it safe" The SECRET_KEY file should never be shared or made public.
You will also need to install the python packages required for your particular database backend:
MySQL
pip3 install mysqlclient
Postgresql
pip3 install psycopg2
Configure InvenTree Options
Once the required software packages are installed, the database and server options must be configured.
Edit the configuration file at /home/inventree/src/InvenTree/config.yaml.
!!! info "Config Guidelines" Refer to the configuration guidelines for full details.
!!! warning "Configure Database"
Ensure database settings are correctly configured in config.yaml before proceeding to the next step!
Create Database
!!! todo "TODO"
Development Server
The InvenTree development server is useful for testing and configuration - and it may be wholly sufficient for a small-scale installation.
Running on a Local Machine
To run the development server on a local machine, run the command:
inv server -a 127.0.0.1:8000
Serving on the address 127.0.0.1 means that InvenTree will only be available on that computer. The server will be accessible from a web browser on the same computer, but not from any other computers on the local network.
Running on a Local Network
To enable access to the InvenTree server from other computers on a local network, you need to know the IP of the computer running the server. For example, if the server IP address is 192.168.120.1:
inv server -a 192.168.120.1:8000
!!! warning "Not For Production" It should be noted that the development server provided with django / InvenTree is probably not suitable for your production environment. Instead, use a proper web-server (such as Gunicorn, below).
Install Supervisor
!!! todo "More details here"
Install the supervisor process manager:
sudo apt-get install supervisor
Gunicorn
Following is a simple tutorial on serving InvenTree using Gunicorn. Gunicorn is a Python WSGI server which provides a multi-worker server which is well suited to handling multiple simultaneous requests. Gunicorn is a solid choice for a production server which is easy to configure and performs well in a multi-user environment.
Install Gunicorn
Gunicorn can be installed using PIP:
pip3 install gunicorn
!!! warning "Python Environment" Ensure that gunicorn is installed within the same python environment context as the InvenTree install - otherwise gunicorn will not be able to import the correct python modules.
Configure Static Directories
Directories for storing media files and static files should be specified in the config.yaml configuration file. These directories are the MEDIA_ROOT and STATIC_ROOT paths required by the Django app. Ensure that both of these directories are correctly configured for your setup.
Collect Static Files
The required static files must be collected into the specified STATIC_ROOT directory:
inv static
This command collects all of the required static files (including script and css files) into the specified directory ready to be served.
Configure Gunicorn
The Gunicorn server can be configured with a simple configuration file (e.g. python script). An example configuration file is provided in InvenTree/gunicorn.conf.py
{% include 'gunicorn.conf.py' %}
This file can be used to configure the Gunicorn server to match particular requirements.
Run Gunicorn
cd InvenTree
gunicorn -c gunicorn.conf.py InvenTree.wsgi