WebRTC Softphone Connection Drops in Architect Flow Performance View

Just noticed that the conversation detail views within the Performance dashboard are failing to load specific WebRTC softphone sessions for agents located in our Paris office. The issue appears isolated to interactions routed through our primary inbound flow (v4.2.1) after the recent platform update. While the queue activity metrics show a healthy volume of answered interactions, the granular call recordings and transcript data are unavailable for approximately 15% of the sessions logged between 09:00 and 11:00 CET.

The softphone client logs indicate a successful connection establishment, yet the backend reporting module seems to lose the session context shortly after the bridge is established. This discrepancy is causing significant issues with our quality assurance sampling process, as we cannot reconcile the dashboard totals with the actual available recordings.

Error: Session ID [UUID] not found in conversation archive. Check WebRTC signaling logs for timeout events.

The environment is running the latest Genesys Cloud release. Has the team observed similar latency or data synchronization gaps between the WebRTC signaling layer and the Performance analytics engine? We need to determine if this is a configuration error within the flow architecture or a platform-side data retention issue.

If I remember correctly, this often stems from mismatched webrtc_enabled flags in the user profile versus the flow configuration. Check if the data_action triggering the screen pop is timing out on the ServiceNow side, causing the session state to desync. Verify the webhook_payload structure matches the expected schema for digital channels to ensure the recording link is properly attached.