I'm currently facing ingestion errors in Amazon QuickSight due to dataset timeouts during SPICE ingestion. Unfortunately, I can't change the infrastructure that's responsible for triggering the ingestion jobs. To mitigate this, I'd like to identify which datasets are taking the most time to upload, so I can schedule their refreshes during off-peak hours. I’ve already checked the list_ingestions and describe_ingestion API calls, as well as list_refresh_schedules, but none of them seem to provide the actual ingestion duration (start and end timestamps, or a time delta). Is there a way—either via API or CloudWatch logs/metrics—to track how long each ingestion is taking? Any help or workaround would be appreciated! Thanks in advance. Continue reading...
