Video recording

WebSummoner can capture the browser screen of any session and save it as an MPEG-4 (H.264) file. Recording works by attaching a small ffmpeg-based container to the browser container — so it works with any browser image, but only when browsers run in containers.

The default recorder image (websummoner/video-recorder:latest-release) is used automatically; pull it once on every host:

docker pull websummoner/video-recorder:latest-release

Enabling recording

Add the enableVideo capability to the sessions you want recorded (all the naming and tuning options are in the capabilities reference):

{ "enableVideo": true, "videoName": "checkout-flow.mp4" }

videoName must be a plain file name — no folders.

Where files are stored

• WebSummoner as a binary — ./video (override with -video-output-dir).

• WebSummoner in Docker — mount a host directory at /opt/websummoner/video and set OVERRIDE_VIDEO_OUTPUT_DIR to the same host path, so the recorder container writes to your storage:

  docker run -d --name websummoner \
    -p 4444:4444 \
    -v /var/run/docker.sock:/var/run/docker.sock \
    -v $PWD/config/:/etc/websummoner/:ro \
    -v $PWD/video/:/opt/websummoner/video/ \
    -e OVERRIDE_VIDEO_OUTPUT_DIR=$PWD/video/ \
    websummoner/websummoner:latest-release

  Why both the volume and the env var? The browser container writes the video, but it must land on your disk — see bind mounts for how host/container paths relate.

Tip

Configuration Manager sets all of this up automatically with cm websummoner start.

Downloading videos

Direct link (available after the session finishes — the temporary file is renamed to <session-id>.mp4 at session close):

http://websummoner-host.example.com:4444/video/<filename>.mp4

List all files:

http://websummoner-host.example.com:4444/video/

Deleting videos

WebSummoner intentionally has no automatic cleanup — pick one:

Scheduled deletion — remove files older than 2 hours (Unix):

find /path/to/video/dir -mindepth 1 -maxdepth 1 -mmin +120 -name '*.mp4' | xargs rm -rf

HTTP API — delete from passed tests, for example:

curl -X DELETE http://websummoner-host.example.com:4444/video/<filename>.mp4

Recording audio

The standard WebSummoner browser images run a PulseAudio server inside every container. When the recorder can reach it (it connects over the container network using a shared cookie), the mp4 automatically gets an AAC audio track alongside the video — no flags or capabilities needed.

This means a session that plays media (a video site, an audio file, a WebRTC call) is captured with sound. To opt out — for privacy, CPU or file size reasons — set the enableAudio: false capability and the recording will be video-only.

To verify a recording contains audio:

ffprobe -show_entries stream=codec_type video.mp4   # look for codec_type=audio
ffmpeg -i video.mp4 -map 0:a -af volumedetect -f null -   # loudness stats

Note

Browsers block autoplay with sound until a user gesture. To let tests start playback programmatically, pass the matching browser flag, e.g. for Chrome: goog:chromeOptions: {"args": ["--autoplay-policy=no-user-gesture-required"]}.

Choosing a codec

The default libx264 gives the best compatibility. If recording consumes too much CPU, switch per session with the videoCodec capability (e.g. "mpeg4"). List everything your recorder supports:

docker run -it --rm --entrypoint /usr/bin/ffmpeg websummoner/video-recorder -codecs

On Windows

Use a named volume or a converted path for the mounts, and keep OVERRIDE_VIDEO_OUTPUT_DIR identical to the mounted host path:

docker volume create websummoner-videos
$current = $PWD -replace "\\", "/" -replace "C", "c"
docker run -d --name websummoner `
  -p 4444:4444 `
  -v //var/run/docker.sock:/var/run/docker.sock `
  -v ${current}/config/:/etc/websummoner/:ro `
  -v /c/websummoner/video/:/opt/websummoner/video/ `
  -e OVERRIDE_VIDEO_OUTPUT_DIR=/c/websummoner/video/ `
  websummoner/websummoner:latest-release