Skip to content

Latest commit

 

History

History
92 lines (66 loc) · 3.29 KB

README.md

File metadata and controls

92 lines (66 loc) · 3.29 KB
Raw

Nebula worker

GitHub release (latest by date) Maintenance Last commit Python version

This repository contains the code for Nebula 6 worker node. The worker node is responsible for processing media files, handling metadata, automating transcoding, and controlling playout servers.

This worker node can be run in multiple instances to handle the increasing demand for processing and automation of media files. It is a crucial component of the Nebula MAM, providing a scalable and efficient solution for processing large amounts of data.

The worker node code is constantly being updated and improved to provide the best performance and reliability. If you have any suggestions or encounter any issues, feel free to open an issue in the repository or contribute to the code by submitting a pull request.

Installation

Docker

Build the image locally from the included Dockerfile or pull nebulabroadcast/nebula-worker from Docker Hub.

Docker compose is usually the best way to run Nebula worker node, since all the configuration is done in a single docker-compose.yml file.

In order to mount external storage, the container need to be privileged. By running the container in privileged mode, it will have access to the host's resources, including the ability to mount external storage.

The following command will run the container in privileged mode. Keep in mind that worker node services are started based on the hostnames, so setting the container hostname is crucial.

In case worker is running a service which expects incoming network traffic, such as play service, don't forget to expose the ports.

version: '3.7'

worker:
  image: nebulabroadcast/nebula-worker:latest
  hostname: worker01
  cap_add:
    - SYS_ADMIN
    - DAC_READ_SEARCH
  privileged: true
  ports:
    - 42100:42100    # play service
    - 6250:6250/udp  # OSC control for casparcg

Bare metal

Nebula is developed and tested on Debian Buster. Other distributions may work.

Software requirements:

  • Python 3.10+
  • Poetry
  • FFMpeg
  • cifs-utils (for mounting SMB shares)
  • mediainfo (used by import service)
  • melt (optional)

Installation:

  1. Install the required software
  2. Clone this repository to /opt/nebula
  3. Install dependencies with poetry install
  4. Run the worker with poetry run python -m dispatch

Nebula worker must run as root in order to mount external storages.

Configuration

The worker node can be configured using environment variables. See the nebula/config.py file for a list of available options and their defaults.

Most important directives are:

  • NEBULA_SITE_NAME
  • NEBULA_POSTGRES - PostgreSQL connection string (e.g. postgresql://user:password@host:port/database)
  • NEBULA_REDIS - Redis connection string (e.g. redis://user:password@host:port/database)
  • NEBULA_PLUGIN_DIR - Path to the directory containing plugins

When installing on bare metal, you may use .env file for settings enviroment variables.