Just noticed that our integration is failing when attempting to fetch screen recording metadata for a high-volume call batch.
The environment is a Genesys Cloud Premium App operating across three distinct organizations. We are using the standard OAuth 2.0 client credentials flow for service-to-service authentication.
The specific endpoint triggering the failure is GET /api/v2/recordings/screen-recordings.
We are passing a filter query to retrieve recordings from the last 24 hours. The payload includes the standard headers and a valid access token.
The response is a 500 Internal Server Error. The error body contains a generic message with no specific stack trace. It just says “An internal error occurred.”
This behavior started appearing after the latest platform update last Tuesday. Previous runs of the same job completed successfully.
We have verified that the OAuth token has the necessary scopes. The scopes include recordings:read and screen-recordings:read.
The issue persists across all three organizations. It is not isolated to a single tenant.
We are seeing a pattern where the error occurs when the result set exceeds approximately 500 records. Smaller batches return data correctly.
Pagination is handled via the standard cursor mechanism. We are not hitting any obvious rate limits. The request rate is well below the documented threshold.
Has anyone else encountered this specific 500 error with screen recordings?
We are considering switching to a polling mechanism with smaller batch sizes as a workaround. However, this impacts our SLA for transcript processing.
The Architect flow triggering the retrieval is simple. It uses a REST API step to call our backend. The backend then queries the Genesys Cloud API.
The timeout settings in Architect are default. The issue is clearly on the API side, not the flow execution.
Any insights into recent changes to the screen recording API would be helpful. We need a reliable way to export this data for our analytics dashboard.