Skip to content

Overview

Handle missing data in time-series queries

Most analytics start by rolling raw events into fixed windows with time_bucket(). That works until a window is empty because sampling was irregular or a collector failed. Gapfilling inserts placeholder rows so timelines stay contiguous; you then interpolate or carry values forward as needed.