Heartbeat.
also: heartbeats · job heartbeat · job heartbeatsA periodic 'still alive' signal; when heartbeats stop arriving, the node or job is presumed dead and its work reassigned.
the two-minute refresher
- Long-running jobs update a last_heartbeat timestamp as they progress.
- A detector finds jobs whose heartbeat is older than the threshold and requeues them.
- Missing heartbeats can mean slow, not dead: re-run work must be idempotent or fenced.