One console for your whole stack
Servers, websites, uptime checks, AI agents, gateways, SSL certificates and databases — SiteSonar watches them all the same way, on your own infrastructure. Push a number from anything and it becomes a live, alert-ready widget.
Every layer, watched the same way
From the metal up to your AI workloads, each entity follows the same model — collect samples, evaluate tolerances, open and close incidents. Mix and match exactly what you need.
Servers & system
CPU, load average, memory, swap and disk usage from a tiny self-updating push agent — on Linux or Windows. A heartbeat URL covers boxes you can't install on.
Websites & uptime
Concurrent HTTP checks for up/down, response time and status codes, with a public status page for the sites your users care about.
Active checks
Probe anything reachable: TCP, UDP, ICMP ping, DNS resolution and blacklist lookups — plus passive callback pings reported by your own jobs.
AI agents & gateways
Poll an API or AI gateway for up, latency and status, and flatten its JSON or Prometheus /metrics endpoint into widgets. Track LLM tokens, cost, latency and error rate.
SSL certificates
Track days-to-expiry plus chain and hostname validity for any domain, and get warned well before a certificate lapses.
Databases
Watch MySQL throughput, connection pressure and the slow queries that are actually hurting you, so the offending instance is easy to spot.
Apache
ApacheWatch follows mod_status workers, request rates and the error log — including abusive-IP detection, with per-domain log configuration.
File watch
Confirm a file exists and keep an eye on its size, age and contents — perfect for backups, exports and queue spool files.
Custom & web analytics
Push any numeric metric from any language or device — each auto-appears as a widget. Plus per-site web-traffic analytics from GA4 and Search Console.
Zero-config metrics, tolerances you set once
Send a number and SiteSonar does the rest. The generalized agent layer turns whatever you push into live gauges and sparklines — no schema, no dashboards to wire up.
- Auto-discovery. Every numeric metric registers itself on first push and picks a sensible gauge or sparkline widget automatically.
- Per-metric tolerances. Set warn and critical thresholds and a direction for each metric — no global rules to fight with.
- Confirm-after-N debounce. A breach must persist across consecutive checks before it pages you, so a single blip never wakes anyone.
- Heartbeat dead-man's switch. A /ping URL any cron or service hits on its own schedule — stop checking in and the agent flips to offline. No daemon required.
- Self-updating agents. Agents validate and replace themselves on a successful push, so a version bump rolls out across the fleet on its own.
Alerting that doesn't cry wolf
Get told the moment something goes dark — and only then. Debounced breaches, clean incident timelines and the channels your team already lives in.
Multi-channel alerts
Route notifications to email, SMS, Pushbullet, Twitter, Pushover and Telegram — pick the channels that reach the right people.
Confirm-after-N debounce
Require a metric to breach across N consecutive checks before it pages you, so transient spikes stay quiet.
Incident timelines
Incidents and events open and close automatically with a full timeline, and reminders repeat while something is still unresolved.
Per-metric tolerances
Tune warn and critical thresholds and the comparison direction for each individual metric, right from the console.
Heartbeat dead-man's switch
A simple /ping endpoint flips an agent to offline the moment it stops checking in — catch silent failures with no daemon to install.
Light/dark Fleet Console
A responsive single-pane dashboard with server-rendered SVG gauges and sparklines that paint instantly in light or dark.