Skip to content
This repository has been archived by the owner on Jul 17, 2024. It is now read-only.

Latest commit

 

History

History
94 lines (66 loc) · 4.57 KB

README.md

File metadata and controls

94 lines (66 loc) · 4.57 KB
Raw

Videoroom

This project is archieved and ongoing development was moved to organization fishjam-cloud

Videoroom is an open-source, basic video conferencing platform using WebRTC. It is based on fishjam, a general-purpose media server. Videoroom may be a good starting point for building your own real-time communication solution using Elixir and Fishjam.

Running with Docker

The simplest way to run videoroom is with use of Docker. To do so, modify .env.example file by setting the EXTERNAL_IP to your private IP address (this can't be loopback!) and type:

docker compose --env-file .env.example up

Running from source

Make sure to have installed Elixir and npm first.

Running the Videoroom requires connecting to an instance of Fishjam Server.

When running locally, you can start an instance of Fishjam inside docker using docker compose.

EXTERNAL_IP=<your ip in local network> docker compose -f docker-compose-dev.yaml up

Now you can start the Videoroom backend:

  • Run mix setup to install and setup dependencies
  • Start Phoenix server with mix phx.server

When running the build version of the Phoenix app, you must specify the addresses of the Fishjam and backend service. As well as the authentication token via the environment variables:

BE_JF_ADDRESS=<IP_ADDRESS>:<PORT1> OR <DOMAIN1> #Example `127.0.0.1:5002 OR room.fishjam.ovh`, if not provided in dev environment `localhost:5002` is used.
BE_JF_SERVER_API_TOKEN=<TOKEN> #This must be the same token that was setup in fishjam. In `docker-compose-dev.yaml` we setup `development` and this variable is used by default in `dev` environment

Optionally, in production, these variables can be set:

  • BE_PEER_JOIN_TIMEOUT - can be used to limit the period in which a new peer must join the meeting,
  • BE_JF_SECURE_CONNECTION - enforces connecting the backend to fishjam through wss protocol,
  • BE_HOST - address of backend
  • JF_CHECK_ORIGIN - define whether fishjam should check origin of incoming requests

Next you have to start a Videoroom frontend:

  • Run npm ci --prefix=assets to install and setup dependencies
  • Start vite server with npm run dev --prefix=assets

Now you have all needed components to use fishjam videoroom.

Production

To run the Videoroom in production on one machine you can use the provided Docker compose file and adjust it to your needs. Example configuration is provided in docker-compose.yaml and .env.example files. You can copy the .env.example file to .env and adjust it to your needs.

Deployment with load-balancing

docker-compose.yaml allows to run a fishjam videoroom with multiple fishjams but all of that runs on the same machine. For properly using load-balancing two machines will be needed and docker-compose-deploy.yaml will be used. You can see our deployment workflow here. This deployment is pretty simple. All containers besides fishjam2 are started on node1 and fishjam2 is started on node2. All environment variables used in our deployment are presented below:

DOMAIN=<FRONTEND_DOMAIN>
JF_SERVER_API_TOKEN=<API_TOKEN> #The same API token is used for all fishjams
BE_JF_ADDRESS=<DOMAIN_FISHJAM1> OR <FISHJAM_IP>:<FISHJAM_PORT>  #Used by backend to create a notifier and to communicate with fishjam
PROMETHEUS_TARGETS=<FISHJAM_IP>:9568 #Addresses on which prometheus will query for data
BE_HOST=<BACKEND_DOMAIN>
GF_SECURITY_ADMIN_PASSWORD=<GRAFANA_PASSWORD>
GF_SECURITY_ADMIN_USER=<GRAFANA_LOGIN>
BE_JF_SECURE_CONNECTION=true
JF_CHECK_ORIGIN=false

Tests

We use Divo in tests, which is responsible for starting docker containers.

When running locally run tests using mix test, which starts Fishjam in a container.

On CI both Fishjam and the tests are run inside docker. If needed, e.g. when the tests are failing on the CI, but not locally you can simulate those conditions with mix integration_test.

Copyright and License

Copyright 2020, Software Mansion

Software Mansion

Licensed under the Apache License, Version 2.0