Data timeout

When Flinn requests data from an upstream source — a literature database, a vigilance database, a regulatory portal — the connection has a fixed time budget. If the upstream response is too slow, the request can time out. The user-visible effect is an incomplete result or an interrupted action.

Hazardous situation: User-initiated actions fail to complete; data acquisition is incomplete or incorrect because of a timeout against an upstream source.

• Retries against transient upstream slowness. Most timeouts are short-lived and the underlying integration retries automatically. Background on the integrations: From Manual to Automated: Flinn's Search Advantage Explained, Pubmed, Cochrane Reviews, ICTRP – Clinical Trials, Google Scholar best practices.
• Refresh and re-execute. If a search returns incomplete results because of a timeout, re-running it a few minutes later usually returns the full dataset; see Synchronization issues with external database for the user-facing workflow.
• Smaller scopes when needed. Very broad queries are more likely to hit timeouts; narrower scopes are more robust. See Define a Search Strategy and Understanding Operators.
• Cross-check critical evaluations. Validate against the source database for high-stakes evaluations; see Why do I find different results when I compare the official database and Flinn?.
• Report sustained timeouts. If the timeout pattern persists, Report a problem or a bug. Related: External database downtime and Request overload.