Skip to main content

Dashboard and Control Room

Dashboard & Control Room

Overview

The Probr dashboard provides a centralized control room to monitor the health of all your clients, sites, and probes in real time. The main endpoint returns a hierarchical view:

Client → Sites → Probes → Latest results

Each level aggregates the status of its children using a worst-status priority system.

Status Hierarchy

Probr uses four status levels, ordered from most severe to least:

Status Priority Meaning
critical Highest A probe detected a major failure
error High The probe execution itself failed
warning Medium A threshold has been breached
ok Normal Everything is working correctly

A site's overall status is the worst status among its active probes. A client's overall status is the worst status among its active sites.

API Endpoint

GET /api/dashboard/overview

Returns the full control room view with all active clients, their sites, probe statuses, and recent alerts.

Response schema:

{
  "total_clients": 3,
  "total_sites": 8,
  "total_ok": 6,
  "total_warning": 1,
  "total_critical": 1,
  "clients": [
    {
      "client_id": "uuid",
      "client_name": "Acme Corp",
      "overall_status": "warning",
      "total_sites": 2,
      "sites_ok": 1,
      "sites_warning": 1,
      "sites_critical": 0,
      "active_alerts": 1,
      "sites": [
        {
          "site_id": "uuid",
          "site_name": "acme.com - Production",
          "site_url": "https://acme.com",
          "overall_status": "ok",
          "probes": [
            {
              "probe_type": "http_health",
              "status": "ok",
              "message": "HTTP 200 in 142ms",
              "last_check": "2025-02-24T10:30:00Z",
              "response_time_ms": 142.5
            }
          ],
          "active_alerts": 0
        }
      ]
    }
  ],
  "recent_alerts": [
    {
      "id": "uuid",
      "site_id": "uuid",
      "probe_config_id": "uuid",
      "severity": "warning",
      "probe_type": "http_health",
      "title": "[WARNING] http_health — staging.acme.com",
      "message": "HTTP 503 — Service Unavailable",
      "is_resolved": false,
      "resolved_at": null,
      "notified_at": "2025-02-24T10:31:00Z",
      "created_at": "2025-02-24T10:30:00Z"
    }
  ]
}

Response Fields

Field Type Description
total_clients int Number of active clients
total_sites int Number of active sites across all clients
total_ok / total_warning / total_critical int Site count by status
clients[] array List of client overviews
clients[].sites[] array List of site overviews for this client
clients[].sites[].probes[] array Latest result for each active probe
recent_alerts array Last 20 alerts (resolved and unresolved)

How It Works

  1. The endpoint loads all active clients with their sites and probe configurations
  2. For each active probe, it fetches the latest result from the database
  3. It counts unresolved alerts per site
  4. It computes aggregated statuses bottom-up (probe → site → client)
  5. It returns the last 20 recent alerts across all sites

Use Cases

  • At a glance monitoring: see which clients/sites need attention
  • Drill down: identify which specific probe triggered an issue
  • Alert feed: review recent alerts and their resolution status
  • SLA reporting: track uptime across your client portfolio
Back to top