An analyst runs a KQL query against an Eventhouse table holding vibration sensor readings from a vault, filtering for the last five minutes, and gets back an unexpectedly empty result. The Eventhouse itself reports as healthy, and querying for the last hour instead returns plenty of rows, just none in the most recent few minutes. What is the most likely explanation?

Ingestion into an Eventhouse table involves a pipeline of buffering and batching before data becomes queryable, so there is normally a small delay between an event being produced and it appearing in query results; a query scoped to only the most recent few minutes can land inside that gap and return nothing even though ingestion is working correctly and data will appear moments later. The fact that widening the window to the last hour returns rows confirms the pipeline is healthy overall and that the narrow window simply caught the freshest edge of arriving data before it landed. A syntax error would typically fail the query outright or misbehave consistently regardless of the time range chosen, not selectively return zero rows only for the newest slice of time. A dropped and recreated table would produce a schema or object-not-found style error, and would also lose the older data the wider query is clearly still returning. Sensors going silent would show as a growing gap that persists even in wider queries over time, not a normal-looking hour of data with just the last few minutes missing. The concrete check is to rerun the same narrow query a minute or two later and confirm the previously missing rows have since appeared, which distinguishes latency from a real ingestion break.