APScheduler Missed Run Alert

Notice when an APScheduler job SILENTLY STOPPED FIRING -- which APScheduler will not tell you.

A liveness layer over your OWN run records: give it the schedule, the misfire grace time, an injected clock, and what actually ran from your own store, and it tells you which nominal runs were EXPECTED, which were OBSERVED, and which are MISSING -- with the reason, where one can honestly be determined. Zero runtime dependencies; apscheduler is never imported, only duck-typed via its own public trigger protocol.

For APScheduler -- not affiliated with or endorsed by its authors, never forked, vendored or redistributed.

THE SILENT FAILURE. A job that does not run raises nothing. There is no exception, because no code executed. The absence of a log line is indistinguishable from a quiet week, and every monitoring system that alerts on errors is looking for something that by definition never happened.

The causes are ordinary: the run was skipped past its misfire grace time, several pending runs were coalesced into one, a max_instances limit declined it, or the scheduler process died and came back with no memory of what it owed. Each produces a gap that nothing announces.

IT NAMES THE REASON where one can honestly be determined, and says NOT DETERMINABLE where it cannot -- because a missing run attributed to the wrong cause sends someone to fix a grace time when the process was dying.

VERIFIED: 38 tests, measured by running the suite.

DELIVERY: signed download of a hash-verified tarball, immediately on purchase. Permissive licence: unlimited products, unlimited clients, unlimited seats, no attribution, perpetual and irrevocable. One restriction, do not republish the source as source.