Quick question, has anyone seen this weird error? with our BYOC trunk failover logic affecting AI Bot interactions. When the primary carrier drops, the secondary trunk picks up after a 200ms delay, causing the SIP 408 Request Timeout in the Architect flow. This breaks the voice-to-text transcription pipeline. Our trunks are in Singapore, so latency shouldn’t be this high. Has anyone seen SIP re-registration delays disrupt bot handoffs specifically during carrier switchover?
Has anyone seen SIP re-registration delays disrupt bot handoffs specifically during carrier switchover?
You need to adjust the JMeter ramp-up to simulate realistic SIP retry intervals. The 200ms gap is too tight for WebSocket reconnection limits. Try increasing the sip.timeout in your load test config to 500ms. This mimics actual trunk failover behavior and prevents the 408 timeout from breaking the transcription pipeline.