Status Harbor Docs
Status Harbor watches services behind a firewall. The Lighthouse agent runs inside your network, opens only outbound HTTPS, and probes internal APIs, private databases and Kubernetes services from where they actually live. No VPN, no inbound ports.
The same product also monitors public endpoints from regional probes
- HTTP, TCP, UDP, SSL or DNS - with one shared dashboard and incident pipeline. Alerts go to Slack, Telegram or email. Webhook and WhatsApp are on the roadmap.
Beyond uptime, every Lighthouse agent also reports host and cluster metrics: CPU, memory, disk and network from the host it runs on, plus per-node Kubernetes capacity when you deploy the DaemonSet flavour of the chart. Threshold alerts on those metrics fire into the same incident pipeline as monitor up / down events, so a CPU breach reaches the same Slack channel as a failed health check.
Start here
- Quickstart - your first monitor in under a minute.
- Monitors - HTTP, TCP, UDP, SSL and DNS checks.
- Cron checks - heartbeat monitoring for scheduled jobs.
- Notifications - Slack, Telegram and email.
- Lighthouse - monitor private-network services.
- Host & cluster metrics - CPU, memory, disk, network and Kubernetes capacity with threshold alerts.
- Terraform provider - manage Lighthouses as code.
How it works
The dashboard at console.statusharbor.io runs the control plane: monitors, schedules, incident state, metrics and alerts.
Public monitors run from Status Harbor's regional probes (United States, Germany and Singapore - see Regions). Lighthouse-bound monitors run from your agent. Both feed the same incident pipeline: a single failure opens an incident, a single success resolves it.
Host and cluster metrics ride on the same Lighthouse agents and the same outbound HTTPS connection - no second collector to install. The Metrics workspace is where charts, threshold rules, snoozes and firing alerts all live; warning and critical breaches join the same incident pipeline as your monitor alerts.