Vendor Status Dashboard

Live check

Is Datadog down right now?

Checking Datadog’s status page…

Fetched live from status.datadoghq.com by your own browser — we never proxy, cache, or fake it.

While you wait: the Datadog incident checklist

  1. Confirm which site you're on. Datadog runs independent regional sites (US1, US3, US5, EU1, AP1 and newer sites). Incidents are usually site-specific — check the status page for your site (EU orgs: status.datadoghq.eu) before assuming you're affected.
  2. Assume you're partially blind. If intake or monitor evaluation is degraded, alerts may simply not fire. Stand up an independent signal on your crown-jewel services now: another vendor's uptime checks, raw health endpoints, or a watch loop from a shell.
  3. Leave the agents alone. Don't mass-restart agents — they retry on their own, and a restart storm adds churn without fixing anything upstream. Expect gaps or backfilled data around the window rather than clean lines.
  4. Brace for the recovery storm. Delayed monitor evaluations and recovering synthetics can fire late, duplicated, or flapping when ingestion catches up. Warn whoever is on call before it happens, not after.
  5. Annotate the gap. Record the incident window wherever your SLO data lives: missing data isn't zero data, and next week someone will read that dashboard dip as a real regression.

What breaks when Datadog has issues

A Datadog incident is a monitoring blackout, not a production outage: your services keep running, but dashboards go stale, monitors stop evaluating, and pages that should fire may never reach PagerDuty or Slack. The dangerous failure mode is silence — 'no alerts' reads as 'all good' at exactly the moment it isn't. Synthetics, log ingestion and APM traces can lag or gap, and any automation keyed off Datadog monitors — auto-rollbacks, scaling webhooks — is now acting on stale signals.

FAQ

Will my monitors still alert during a Datadog outage?

Treat alerting as degraded. If intake or monitor evaluation is affected, alerts can be late or missed entirely. Watch independent signals for your critical paths until the incident resolves.

Is my metrics data lost?

Agents are designed to buffer and retry through short interruptions, so recent data often backfills after recovery — but expect gaps around longer windows. Either way, don't read the gap as zeros.

The status page shows an incident but my org seems fine — why?

Datadog operates separate regional sites and incidents are usually scoped to one. Make sure you're reading the status for the site your org lives on — check your app URL (app.datadoghq.com, app.datadoghq.eu, us3/us5 variants).

Does a Datadog outage put production at risk?

Not directly — the agent fails quietly and your workloads don't depend on it. The risk is operating blind, plus any automation that consumes Datadog signals.

Get pinged in your toolbar the moment Datadog’s status changes

Vendor Status Dashboard is a Chrome extension for on-call engineers: Datadog and every other vendor you depend on, one badge, answer in five seconds. Free for 5 vendors, no account.

Join the waitlist — launching soon

Also on your stack?

Not affiliated with Datadog. Status data comes from their public status page.