ObsrvHQ Documentation

Everything you need to connect ObsrvHQ to your existing alerting stack, configure suppression rules, and understand why each alert was suppressed or paged.

Active setup takes about 5 minutes. The 48-hour window that follows is the baseline learning period running in the background. Your alertmanager stays unchanged and all alerts pass through until the window completes.

How ObsrvHQ works

ObsrvHQ is a filter layer between your metrics source and your alertmanager. It reads your existing alert rules, builds per-series dynamic baseline envelopes from observed time-series data, and makes a suppress-or-page decision before anything reaches your on-call team.

The three-step flow:

  1. Connect: Point ObsrvHQ at your Prometheus endpoint or Datadog API. One YAML config block. Active setup takes under 5 minutes.
  2. Learn: ObsrvHQ watches every metric series for 48 hours, building per-hour, per-weekday baseline envelopes. During the learning window, all alerts pass through unchanged. No ML training required from you.
  3. Filter: Alerts that fall within the learned envelope for their time window are suppressed. Genuine deviations outside the envelope page your team. Suppression decisions include the reason: which envelope bound was exceeded, by how much, and the configured sensitivity at the time.

Authentication

All API requests require a bearer token. Generate your API key from your workspace settings after signing up. Treat the API key as a secret: it has read-write access to your workspace configuration and decision history.

curl -H "Authorization: Bearer YOUR_API_KEY" \
  https://api.obsrvhq.com/v1/baselines

Support

Questions not answered here? Email [email protected]. We respond within one business day.