Awesome

moNNT.py Async Usenet Server

moNNT.py is an async usenet server written in Python. To run the server, cd into its directory and then

$ poetry run python main.py

This software was written as part of my Bachelor thesis (Computer Science). It should therefore not be regarded as a full-fledged replacement for INN or other production news servers. As of right now, it is a proof-of-concept that demonstrates the operation of a news server in a disruption-tolerant network (DTN) environment. The DTN connectivity is handled by the most excellent dtn7-rs DTN network daemon.

For more information about how to run the server in conjunction with the dtnd, see the backend README.

Features

• async operation for handling multiple clients simultaneously
• fully exchangeable synchronization and storage backends (see section below)
• pyproject.toml based configuration

Configuration

The server is configured through the flag set in the monntpy.env setting in pyproject.toml. This controls which configuration is loaded. In the server config[dev|prod|test].toml environment specific settings can be configured.

The backends must implement their own configuration scheme. For the DTN7SQLite backend, there is a config.toml in the main backend directory.

Exchangeable Synchronization and Storage Backends

New backends for synchronization and storage can easily be implemented by abstracting from the backend.base.Backend class.

Development Environment

Development and deployment is facilitated by the Poetry dependency manager, which also allows separate development dependencies. Pre-commit hooks can be activated and are managed through the .pre-commit-config.yaml configuration file. Presently, the tools black, isort and flake8 are used to lint and check the source code before committing.

Dockerized Deployment

moNNT.py can easily be run in a Docker container. As a reference setup, a Dockerfile is included in this repo. It is based on the official Python 3.10.7 image python:3.10-slim from Dockerhub.

A few notes on this: Because the current state of the main directory is copied into the image, setup everything as you wish to encounter it in the container. This includes settings in the .env files and backend specific configurations, but also in the Dockerfile and docker/entrypoint.sh as this will control the execution of your server.

Then, in directory of the Dockerfile, run

$ docker build . --tag monntpy

You can then simply run the application with

$ docker run -rm -ti --network="host" monntpy

and connect to it on port 1190.