- Genesys Cloud Release: 2024-02 (Patch 4)
- BYOC Edge Version: 10.1.2
- Environment: Private Cloud (AWS us-east-1)
- Timezone: America/Chicago
- SDK: Genesys Cloud Web SDK 3.15.0
Why does this setting in the WebRTC configuration cause immediate softphone disconnections when the WFM Schedule Publication API is triggered? We have an automated workflow built in Architect that pushes finalized schedules to Genesys Cloud every Tuesday at 07:30 CST. This process works flawlessly for 90% of our agents. However, agents assigned to specific skill groups (specifically those with high adherence requirements) experience a sudden drop in their softphone audio stream exactly when the /api/v2/wfm/schedules endpoint returns a 200 OK.
The error log in the WebRTC client shows a ‘Remote peer closed connection’ event followed by a ‘Signaling error: 1006’. The interesting part is that this only happens if the agent is currently in a ‘Ready’ status in WFM at the moment of publication. If they are in ‘Away’ or ‘Break’, the connection holds steady. We suspect the schedule update is triggering a re-evaluation of the agent’s routing profile, which forces a renegotiation of the WebRTC session. Since the BYOC edge node handles the signaling, there seems to be a race condition where the new schedule constraints are applied before the WebRTC session can gracefully handle the profile update.
We have checked the edge node logs and see no 502 or 504 errors. The latency between the Genesys Cloud core and our edge node is under 15ms. We need to ensure that schedule publications do not disrupt active or ready agent sessions. Is there a specific flag in the WFM API call or a setting in the BYOC edge configuration that should suppress real-time routing profile updates during bulk schedule publishes? We want to maintain shift swap flexibility without breaking the softphone connection for agents who are actively waiting for calls.