Uptime Monitoring
60-second heartbeat monitoring with multi-layer false-alarm prevention.
How It Works
When enabled, Teste.no pings your site every 60 seconds with an HTTP HEAD request (with GET fallback). Pings are staggered across the cycle so sites don't all get checked at the same instant.
Status Indicators
UP
Responding normally
DOWN
Unreachable (after threshold)
DEGRADED
Slow or 4xx responses
UNKNOWN
Not yet checked
False-Alarm Prevention
Three layers protect against false positives:
1
Immediate confirmation retries
When a ping fails, up to 2 retries with 5-second delays — catches transient network blips instantly
2
Consecutive failure threshold
Must fail N checks in a row before marking DOWN (default: 3 cycles ≈ 3 minutes)
3
Degraded confirmation
Must be slow/4xx for N consecutive checks before DEGRADED (default: 2)
What this means in practice: A single network blip won't trigger alerts. A site must be genuinely unreachable for approximately 3 minutes before it transitions to DOWN and notifications fire.
Keyword Verification
Optional but powerful: set a keyword that must appear in the page body. If the keyword is missing — even if the server returns HTTP 200 — the site is marked DOWN.
Server errors
Catches "Something went wrong" pages
CDN failures
Detects stale error pages served by CDN
Database down
Catches DB failures returning maintenance page
Proxy issues
Reverse proxy showing its own default page
Configuration
| Setting | Default | Description |
|---|---|---|
| Enable monitoring | On | Toggle uptime checks for this target |
| Keyword | None | Text that must appear in page body (optional) |
| Failure threshold | 3 | Consecutive failures before DOWN |
| Degraded confirmations | 2 | Consecutive slow/4xx before DEGRADED |
| Ping timeout | 10,000ms | Max wait time per ping request |
| Degraded threshold | 5,000ms | Response time above this = degraded |
What Happens on Downtime
When a target transitions to DOWN:
- A finding and incident are automatically created
- Notifications fire on all configured channels (in-app, SMS, Slack, email)
- Browser alarm sound plays (if enabled)
- Voice alert announces the event via TTS (if configured)
When the site recovers to UP:
- Finding auto-resolves, incident auto-closes
- Recovery notification sent to all channels
- Consecutive failure counter resets
History & Graphs
The target overview shows an Uptime Pulse Monitor with selectable time ranges (24h, 7d, 14d, 30d). Raw checks are kept for 48 hours, then aggregated into daily stats for long-range display up to 90 days.