Skip to content

LIVE DASHBOARD DOCUMENTATION

From first check
to full visibility.

Create a workspace, add a service, and choose how we should reach you.

1. Start monitoring

  1. Create an account and verify your email.
  2. Create a workspace for your company or client. Free workspaces include 5 monitors.
  3. Add a website, DNS record, SSL certificate or heartbeat. Checks run every 1 to 1,440 minutes.
  4. Add your verified email or an HTTPS webhook in Alert destinations, then send a test.

Two consecutive failures open an incident by default. The next passing check resolves it. Maintenance suppresses new notifications while checks continue. Email requires a configured sender; the destination screen reports whether it is connected.

2. Connect heartbeat agents

Create a heartbeat monitor and copy its private URL. The dedicated machine endpoint accepts automated traffic without browser challenges. A heartbeat is overdue after 1.5 times its configured interval. No CPU, memory or disk data is collected.

Linux / macOS

Send a check-in after a successful job, or run this command every minute with cron. Store the URL in a file readable only by the service account.

# /etc/livedashboard/heartbeat.curl (mode 600)
url = "YOUR_PRIVATE_HEARTBEAT_URL"
request = "POST"
fail
silent
show-error
max-time = 15

# crontab: every minute
* * * * * /usr/bin/curl --config /etc/livedashboard/heartbeat.curl

Windows PowerShell

# Store your private URL in an ACL-restricted file.
$url = (Get-Content 'C:\ProgramData\LiveDashboard\heartbeat-url.txt' -Raw).Trim()
Invoke-WebRequest -Uri $url -Method Post -TimeoutSec 15 -UseBasicParsing | Out-Null

Use Task Scheduler to run the PowerShell script every minute. Rotate the heartbeat URL in the monitor details if it is exposed, then update the agent. Heartbeats prove that the agent can reach LiveDashboard; they do not prove that every service on the host is healthy.

3. Use the workspace API

Create a workspace key in Settings. Use https://livedashboard-core.gmtai-growthmode.workers.dev as the machine API base URL and send the key as a Bearer token. It can only access that workspace. Browser sessions use verified identities; viewer members cannot change settings.

GET /api/dashboard?tenant=WORKSPACE_ID
Authorization: Bearer YOUR_WORKSPACE_KEY

POST /api/monitors?tenant=WORKSPACE_ID
Content-Type: application/json
Authorization: Bearer YOUR_WORKSPACE_KEY

{
  "name": "Company website",
  "kind": "http",
  "target": "https://your-company.com",
  "interval_seconds": 300,
  "expected_status_codes": [200],
  "config": { "failure_threshold": 2 }
}

Supported kinds: http, dns, ssl, heartbeat. HTTP monitoring uses GET with optional keyword matching. DNS checks support A, AAAA, CNAME, MX, NS, TXT and CAA. Certificate checks validate TLS and warn 14 days before expiry by default. Targets must use public DNS names. Private IPs and reserved destinations are blocked.

List recent results with GET /api/monitors/ID/checks. Queue a check with POST /api/monitors/ID/check. Pause or resume with PATCH /api/monitors/ID and {"enabled":false}. Include the tenant query parameter on each workspace request.

4. Integrate Nexus IT OS

Open the Monitoring page in Nexus IT OS using your existing Cloudflare Access identity. The Nexus server connects to LiveDashboard through a restricted Cloudflare service binding.

Successful client-directory reads provision active client workspaces idempotently using the Nexus client ID. The monitoring feed includes only Nexus-linked workspaces, monitor status and recent incidents.

Use workspace keys for external automation. Configure an HTTPS webhook for incident and recovery events; event IDs let your receiver deduplicate retries.

Checks, retention & delivery

Checks run from Cloudflare infrastructure with bounded concurrency and request timeouts. We do not offer selectable geographic probe regions. Recent check and delivery history stays available in your workspace. Historical records are preserved during this migration. Dashboard status refreshes every 15 seconds and public pages every 30 seconds.

Alert attempts use exponential retries, up to 6 attempts. A destination is limited to 30 successful alerts per hour. Webhooks may be delivered more than once after a network interruption; use the event ID to deduplicate. An unconfigured mail sender is reported as a delivery error, never as successful delivery.

Free includes 5 monitors. Paid accounts are provisioned after a sales agreement; no payment is collected through this release.