Just noticed that the real-time queue occupancy metrics in the Performance dashboard are failing to update during specific SIP trunk registration cycles. The environment is configured with Genesys Cloud EU-West (Frankfurt) and utilizes three primary SIP trunks for inbound PSTN traffic. The issue manifests when the SIP registrar processes a high volume of REGISTER requests during the shift change at 08:30 CET. Instead of seeing a gradual increase in agent availability, the dashboard reports a temporary drop to zero active agents, followed by a sharp spike that exceeds the actual seat count. This behavior suggests a latency in the state synchronization between the Telephony service and the Performance reporting engine.
The Architect flow handling the initial call distribution is standard, utilizing a ‘Find Agent’ block with skills-based routing. However, when the SIP trunks are re-registering, the ‘Call Connected’ metric in the Conversation Detail view shows a 4-6 second delay before the interaction appears in the queue. During this window, the Performance dashboard calculates occupancy based on stale data, leading to inaccurate KPI reporting for the shift supervisors. The error is not critical enough to drop calls, but it severely impacts the trust in the real-time workforce management tools.
“SIP Registration timeout exceeded for trunk ID: TRK-8842. Retrying in 5 seconds. Impact on real-time dashboard state synchronization observed.”
This message appears in the System Logs under the Telephony category. The question is whether this delay is a known limitation of the current Performance dashboard architecture when handling bulk SIP re-registrations, or if there is a configuration setting in the Telephony integration to throttle the registration bursts. The business requirement is to have accurate real-time visibility into agent status, and currently, the dashboard is misleading during these peak transition periods. Any insights into optimizing the SIP registration process or adjusting the dashboard refresh rate to mitigate this discrepancy would be appreciated. The goal is to ensure that the reported occupancy aligns with the actual agent state, regardless of the underlying SIP signaling load.