README.md

Test Server

Test server based on WebRTC Medooze Media Server for Node.js with web apps allowing testing its features.

Setup

Install project dependencies:

npm install

Add the Medooze media server for Node.js as a dependency in a version you would like to test (or link local one):

npm install medooze-media-server

Run the test server providing public IP address, by which the server can be reached from the machine running web apps:

node index 127.0.0.1

Web apps are available at port 8084, e.g. https://127.0.0.1:8084.

Usage

Bandwidth estimation

App allows to connect browser as a viewer to a media server, after the stream is stopped it shows collected BWE stats.

Automated benchmark tests

There are pytest tests available to run benchmark suite, which tests media server's BWE algorithm in most common scenarios. The output of benchmark tests are bwe and webrtc-internals stats collected, while running scenarios.

To run tests, create python virtual env and install required dependencies by running the following commands in the root of the repo:

python3 -m venv .venv
. .venv/bin/activate
pip install -r test/requirements.txt

Now you can run tests by executing the following command in the root of the repo:

python -m pytest \
--tc_prefix="ssh [email protected]" \
--tc_interface="ens33" \
--twcc_app_url="https://192.168.0.103:8084/twcc" \
--stats_out="/tmp/stats"

Where:

• tc_prefix is the prefix used for tc invocations, to run it on a remote machine, it can be ssh, as in the example, or docker, e.g. docker container exec mediaserver-test-server-1.
• tc_interface is the interface name on a remote machine used by the media server, which will be managed by tc .
• twcc_app_url is the url of twcc app exposed by test server.
• stats_out is the path, where collected stats will be stored (relative or absolute).

Run

pytest --help

to check all available configuration options under Custom options: section.