Quick question about the timestamp alignment in bulk export jobs. We are running a scheduled export via the Genesys Cloud Recording API to an S3 bucket for a legal hold request. The job completes successfully, but the interaction_start_time in the generated manifest file is consistently offset by 3600 seconds compared to the interaction data pulled directly via the Analytics API. This discrepancy is causing issues with our chain-of-custody validation scripts. The environment is set to Europe/London, and the S3 bucket is in eu-west-2. We have verified that the server-side configuration for time zones is correct. The bulk export job uses the standard JSON format for metadata. Is there a known issue with how the Recording API handles daylight saving time transitions in the metadata generation process? Or is this a caching issue within the export service? We need precise timestamps for the discovery phase, and this drift is unacceptable. Any insights on forcing a manual refresh of the metadata cache or a workaround to align the timestamps would be appreciated. The current workaround involves post-processing the manifest files, but this adds significant latency to the export workflow.
“timestamp”: “interaction_start_time_utc”
If I remember correctly, the bulk export defaults to local time while the Analytics API returns UTC. Switching the manifest field to `interaction_start_time_utc` aligns the timestamps and fixes the drift.