You may experience one or both of the following symptoms when using Data Engine:
Data Engine logs you out unexpectedly during a task. Refreshing the tab does not restore the session — you need to close the tab and reopen Data Engine from the nine-dot menu.
Scheduled reports fail to run or complete.
These issues can have two causes: a platform-level outage affecting the DataHub messaging system, or performance pressure on the shared infrastructure during busy periods.
Platform outages
Some instances of these errors are caused by temporary DataHub outages — for example, connectivity issues with the RabbitMQ server that drives the DataHub messaging system.
When this occurs, our hosting team investigates and resolves the issue, which may involve temporarily pausing message flow and restoring RabbitMQ.
If you are experiencing sudden, widespread failures, this may be the cause. These are resolved by our team and do not require any action on your part.
📌Note: If you suspect a platform outage, please raise a support case so our team can investigate.
Shared infrastructure and peak-time queuing
Data Engine runs on a shared infrastructure with a limited number of threads available to process tasks.
During busy periods — typically early morning — a high volume of tasks from all customers are queued for processing. This means we cannot guarantee that scheduled tasks will execute or complete at the exact time they were scheduled for.
If you have a large number of tasks, particularly large tasks, or schedules with a short timeframe between them, this increases the likelihood of delays or failures during peak times.
🤓Tip: To reduce the impact of peak-time queuing, spread your schedules out so that fewer tasks run at the same time, and avoid very short intervals between schedules.
About Data Engine schedules
Data Engine is a data warehouse solution designed to support business intelligence reporting by pulling data from sources, transforming it, and presenting it to Access Analytics. Its primary use case is not real-time reporting.
Schedules can be configured to run as frequently as every 30 minutes, but during busy periods tasks are queued and may not run immediately. We are aware of the challenges this presents for customers who require time-critical, high-frequency schedules and are actively exploring ways to improve the experience.