Documentation / Capture a dump

Capture a dump

The app's phase-1 input is the standard chrome://webrtc-internals dump — it works with any WebRTC service: Google Meet, Microsoft Teams, or your own application.

Export from chrome://webrtc-internals

While a WebRTC call is running in Chrome (or Edge: edge://webrtc-internals), open a new tab and go to chrome://webrtc-internals.
Expand Create Dump at the top of the page.
Click Download the PeerConnection updates and stats data.
A file like webrtc_internals_dump.txt is downloaded — its content is JSON, and that is what you upload.

Open webrtc-internals early. The dump only contains statistics from while the page was open (sampled every second). Opening it before or at the start of the call gives you the full picture, including connection setup (ICE) events.

Good to know

Retention limit: the browser keeps a bounded number of samples, so for very long calls the tail of the call is retained.
Multiple connections: a dump can contain several peer connections (e.g. an SFU call plus a screen share); all of them are parsed into the same session.
Privacy at parse time: TURN credentials are redacted, SDP bodies are dropped, and ICE addresses can be masked. The raw file itself is not retained — only parsed rows.

Upload

Open the app and go to Upload dump (Data section in the sidebar).
Drag & drop the file, or click to select it.
Parsing, normalization, QoE scoring, and daily rollups run automatically; the call appears in every view within seconds.

Re-uploading the same file is detected by SHA-256 content hash — the app points you to the existing session instead of ingesting a duplicate. Failed parses are recorded in the bronze_uploads ledger with the error message.

← Install Next: Screens →