WebRTC troubleshooter provides a set of tests that can be easily run by a user to help diagnose WebRTC related issues. The user can then download a report containing all the gathered information or upload the log and create a temporary link with the report result.
- Microphone
- Audio capture
- Checks the microphone is able to produce 2 seconds of non-silent audio
- Computes peak level and maximum RMS
- Clip detection
- Mono mic detection
- Audio capture
- Camera
- Check WxH resolution
- Checks the camera is able to capture at the requested resolution
- Checks if the frames are frozen or muted/black
- Detects how long to start encode frames
- Reports encode time and average framerate
- Check supported resolutions
- Lists resolutions from 640x480 to 1920x1080 that appear to be supported
- Check WxH resolution
- Network
- Udp/Tcp
- Verifies it can talk with a turn server with the given protocol
- Udp/Tcp
- Connectivity
- Relay
- Verifies connections can be established between peers through a TURN server
- Host
- Verifies connections can be established between peers with the same IP address
- Relay
- Throughput
- Data throughput
- Establishes a loopback call and tests data channels throughput on the link
- Data throughput
You can run TestRTC against a Xirsys account from the main page:
- Fill in the Xirsys Account panel:
IdentSecret keyChannel
- Leave TURN/STUN URI fields empty to let TestRTC fetch ICE servers from Xirsys.
- Start the test suite as normal.
If custom TURN/STUN URIs are provided, they take precedence over fetched values.
Due to their time duration they are not part of the normal test suite and need to be run explicitly.
- Network latency
- Establishs a loopback call and sends very small packets (via data channels) during 5 minutes plotting them to the user. It can be used to identify issues on the network.
Pull requests and issues welcome! See CONTRIBUTING for instructions. All contributors must sign a contributor license agreement before code can be accepted. Please complete the agreement for an individual or a corporation as appropriate. The Developer's Guide for this repo has more information about code style, structure and validation.
Make sure to install NodeJS and NPM before continuing. Note that we have been mainly been using Posix when developing TestRTC hence developer tools might not work correctly on Windows.
npm installbower updategruntCleans out/ folder if it exists else it's created, then it copies and vulcanizes the resources needed to deploy this on Google App Engine.
grunt build
Run vulcanized version of TestRTC using Google App Engine SDK for Python (requires the Build testrtc step to be performed first).
python dev_appserver.py out/app.yml