Your Server Stopped Reporting.
But Is Everything Actually OK?
Your monitoring agent crashed. Your cron job stopped running. Your health check endpoint returned 200 — but the application behind it was broken. Silence isn't safety.
✓ No credit card · ✓ 5-minute setup · ✓ Works with any server or monitoring stack
Silence looks like health — until it doesn't
The most dangerous server state isn't a red alert — it's no alert at all. When your monitoring stops reporting, everything looks fine. It isn't.
Monitoring agent crashed silently
Your server monitoring agent crashed due to an OOM error. It stopped sending metrics. Your dashboard showed the last known state — 'healthy.' The server was actually under heavy load and nobody knew.
Health check returned 200, app was broken
Your /health endpoint returned 200 — it always did. It just checked that the server was reachable. Behind it, the database connection pool was exhausted and all requests were failing. The 200 masked everything.
Heartbeat cron job stopped running
The cron job that sent your monitoring heartbeat stopped running after a system update changed the crontab. No heartbeat for 8 hours. No alert. You assumed everything was fine because nothing was red.
Dead man's switch monitoring for your infrastructure
NotiLens expects a regular signal from your server. If that signal stops arriving — for any reason — NotiLens fires. The absence of a signal is the signal.
Heartbeat silence detection
Your server sends a regular ping to NotiLens. If the ping stops arriving — for any reason, including a crashed monitoring agent — NotiLens fires. The absence of a signal is the signal.
Deep health check monitoring
Go beyond HTTP 200. Send real application health data — DB connections, queue depth, cache hit rate — and NotiLens monitors it via signal rules and anomaly detection, not just reachability.
Monitoring-the-monitoring
If your primary monitoring tool goes silent, NotiLens catches it. Use NotiLens as a secondary heartbeat layer — your main monitoring sends a periodic ping, and NotiLens alerts if the pings stop.
Manual rules + auto anomaly detection
You get two layers of detection — not one. Set your own thresholds for what you know matters. NotiLens learns the rest automatically.
amount > $100. Fires only when your rule is met.What NotiLens sends when servers go quiet
Every alert fires within minutes of silence detection — with the server name, last check-in time, and silence duration so you can act immediately.
The most dangerous server state is 'quietly broken.'
No errors, no alerts, just silence — and somewhere behind it, something stopped working. NotiLens is designed specifically to detect this: if the expected signal stops arriving, you're alerted.
Silence monitoring catches what threshold monitoring misses
Threshold-only monitoring only fires when values go wrong. Silence monitoring catches when reporting stops entirely. Here's the gap it closes.
heartbeat gap detected — even when the process looks healthy
typical time silent server issues go unnoticed without heartbeat monitoring
to configure a dead man's switch for your infrastructure
Live in 5 minutes
No agents to install. No infrastructure changes. Add a heartbeat cron job and set your silence threshold — your dead man's switch is live.
Create a topic
Create a server-silence topic in NotiLens. This is your dead man's switch.
Set up your heartbeat
Configure a regular ping from your server — cron job, monitoring agent, or health check script.
*/3 * * * * root curl -s -X POST \
https://hook.notilens.com/webhook/TOPIC_ID/send \
-H "X-NOTILENS-KEY: SECRET" \
-H "Content-Type: application/json" \
-d '{"title":"'$HOSTNAME' alive",
"type":"success"}'
Configure silence threshold
Set: "Alert if no heartbeat in 10 minutes." NotiLens fires the moment silence exceeds your window — even if nothing else is red.
More ways to monitor your infrastructure
Server silence monitoring — common questions
Start monitoring server silence
The most dangerous server state is quietly broken. Know within minutes when your server stops reporting. Free 7-day trial.
Start Free — 7-Day Trial