One-click dashboards for popular integrations. Checks which metrics you're already sending, then imports with a single click.
Manage PulseAgent instances enrolled in this workspace.
| Hostname | Version | Last seen | Enrolled |
|---|
HTTP, TCP, and DNS probes run on a cadence from each edge region. Results
land as metrics (pulse_synthetic_up,
pulse_synthetic_duration_seconds) and logs, so you can alert
on them with the rule engine. Editing requires the Admin scope.
Publish a customer-facing status page backed by your existing telemetry.
Components reuse synthetic checks (pulse_synthetic_up) or any
metric series; incidents are auto-derived from firing alerts. The public
page is served unauthenticated at /status (or
/status/<slug>). Editing requires the Admin scope.
Live alert instances generated by your rules — these are what fire, route and notify.
Define alert rules and their inline runbooks. Rules are grouped; each group is evaluated on its own interval. Editing requires the Admin scope.
Silences drop matching alerts during a time window — used for maintenance and known-noise suppression. Active silences take effect on the next evaluation.
The notification pipeline: receivers are delivery targets; the route tree decides who gets paged and how often; inhibitions and mute timings suppress noise. Editing requires the Admin scope.
Users, rotating schedules and escalation policies. Routes reference a policy to page whoever is on call, stepping up after delays until the alert is acknowledged. Editing requires the Admin scope.
Outbound messages produced by routing, delivered with retry. Exhausted messages move to the dead-letter queue, where you can inspect, replay or purge them.
Runbook outcomes aggregated by rule: how many incidents fired, mean time-to-resolve, and which checklist steps get skipped most — so you can tighten the runbooks that aren't working.
This token expires in 30 minutes. Pass it to install.sh or set
enroll_token in agent.toml.