Installation
To quickly test OpenL2M in your environment, run in a docker container! Please see ./docker/test/README.txt
The following steps are for a full, production install of OpenL2M. No containers are supported at this time.
Requirements
OpenL2M has some requirements:
Python 3.10 or greater (OpenL2M is developed and tested on Python 3.11)
net-snmp v5.7 or greater, including net-snmp-devel
the Python “easysnmp” package v0.2.5 or greater.
a web server, with the WSGI capability. We use Nginx in all our documentation. Apache may work but is not tested.
the Django framework, v5.0 or greater.
a PostgreSQL database, running at least version 12. We use v12 and v14 in our testing.
Application Stack Overview
Once installation is complete, you will have the following application stack to get a working OpenL2M application:
Nginx web server
Gunicorn WSGI Process with Python
PostgreSQL database
At the end of this page is an image showing the application stack.
Installation
OpenL2M is developed and tested in a Ubuntu 22.04 LTS environment. All instructions are related to that. However, this should work just fine on other distributions as long as the requirements are met. (We no longer support CentOS Linux v7 and v8 environments, as we no longer developed and tested on them.)
Application Stack Overview