Notification system health
Aggregate health of the notification delivery pipeline. Reports a status verdict (healthy / degraded / unhealthy) computed from the trailing-hour failure rate, plus delivery counters for the last hour and unsubscribe counters for the last 24 hours. Used by uptime monitors and the internal ops dashboard.
Authorizations
Response
OK
Aggregate health snapshot of the notification delivery pipeline.
Returned by GET /health. Used by uptime monitors and the internal
ops dashboard to decide whether the pipeline is keeping up with its
queue. The two time-window buckets are dynamic-shape JSON because the
underlying monitor returns different metrics per window:
last_hour— keysdeliveries(per-status counts) andperformance(failure rate, throughput).last_24h— keysunsubscribes(counts by source).
Aggregate health verdict. Computed from trailing-hour failure rate.
healthy, degraded, unhealthy Delivery counters for the trailing 1-hour window. Keys: deliveries (counts by status), performance (failure rate, throughput).
Activity counters for the trailing 24-hour window. Keys: unsubscribes (counts by source).