Back to Docs

Monitors

HTTP health check monitoring for your endpoints, APIs, and websites.

What are Monitors?

Monitors continuously check your HTTP endpoints to ensure they're responding correctly. When a monitor detects an issue, it automatically creates an incident and sends notifications to your configured integrations.

Each monitor performs health checks at configurable intervals (from 30 seconds to 1 hour) and can validate response status codes, response times, and even response body content.

Creating a Monitor

  1. 1
    Navigate to Monitors

    Click "Monitors" in the sidebar, then click "Create Monitor".

  2. 2
    Configure basic settings

    Enter a name and the URL you want to monitor. Choose GET, POST, PUT, or DELETE as the HTTP method.

  3. 3
    Set check interval

    Choose how often to check your endpoint: 30s, 60s, 5m, 10m, 30m, or 1h.

  4. 4
    Configure validation (optional)

    Set expected status codes, timeout thresholds, and enable SSL certificate monitoring.

Monitor Settings

SettingDescription
NameA descriptive name for your monitor (e.g., "Production API")
URLThe HTTP/HTTPS endpoint to monitor
MethodHTTP method: GET, POST, PUT, DELETE
IntervalHow often to check (30s to 1 hour)
TimeoutMaximum time to wait for a response (default: 30s)
Expected StatusHTTP status codes that indicate success (default: 200-299)
HeadersCustom HTTP headers to send with the request
SSL MonitoringTrack SSL certificate expiration and get alerts before it expires

Understanding Monitor Status

Up

The endpoint is responding correctly and meeting all validation criteria.

Down

The endpoint is failing checks. An incident has been created and notifications sent.

Paused

Monitoring is temporarily disabled. No checks are being performed.

Response Time Metrics

Click on any monitor to see detailed performance metrics:

  • Uptime percentage - How often your endpoint has been available
  • Average response time - Mean response time across all checks
  • P50/P95/P99 percentiles - Response time distribution to identify tail latency
  • Min/Max response time - Range of response times observed
  • Response time chart - Visual timeline of response times with downtime events highlighted

Failure Threshold

To prevent false positives from temporary network issues, monitors require multiple consecutive failures before triggering an incident:

  • Default threshold: 3 consecutive failures
  • Once threshold is reached, an incident is created automatically
  • Recovery is immediate - a single successful check resolves the incident