I’ve spent hours trying to figure out why the Conversational AI analytics endpoint is failing specifically for calls routed through our BYOC trunks.
We are seeing HTTP 500 Internal Server Error responses when querying the /api/v2/analytics/conversations/details/query endpoint with a filter for mediaType: voice and botId associated with our primary customer service bot.
The issue is isolated to traffic entering via our fifteen BYOC trunks in the AP-Southeast-1 region. Calls coming through the standard Genesys Cloud PSTN numbers return data correctly within milliseconds.
The request payload is standard JSON. We are using the Genesys Cloud Java SDK version 1.1.0. The error response body is empty, which makes debugging difficult.
We have verified that the SIP registration status for all trunks is 200 OK. Failover logic is configured correctly, and no trunk is currently in a failed state. Carrier latency metrics are within acceptable thresholds, averaging 45ms.
The problem started appearing after we updated the bot configuration to include a new intent for handling refund requests. This update did not change the underlying routing logic or the BYOC trunk settings.
We suspect there might be a mismatch in how the conversation metadata is tagged for BYOC traffic versus PSTN traffic. The externalContactId field seems to be populated differently for SIP-initiated calls.
Has anyone encountered similar issues with the Bot Analytics API when using BYOC infrastructure?
We need to resolve this quickly as it impacts our quality assurance reporting. The WFM team is relying on this data to evaluate agent-bot handoff efficiency.
Any insights into potential backend processing errors related to SIP trunk metadata would be appreciated. We are currently stuck in a loop of retrying the API call with no success.
Please let me know if you need sample payloads or specific trunk configurations for further analysis.