Uptime monitoring continuously probes a service—such as a web server, API endpoint, or DNS record—at fixed intervals to verify it is running and responding correctly. When the target fails to respond within a timeout threshold, or returns an error, the monitoring system logs the failure and triggers alerts.
Monitoring typically works by sending automated requests to your service and measuring:
- Response time — how long the target takes to reply
- Status code — whether the response is successful (e.g. HTTP 200) or an error
- Uptime percentage — the ratio of successful checks to total checks over a period
Common probe types include HTTP/HTTPS requests to a URL, ICMP ping checks to test network reachability, TCP port checks to verify a service is listening, and DNS record lookups to confirm name resolution. Many monitoring services measure uptime across multiple geographic locations so you can spot regional outages.
Uptime monitoring feeds into SLA (Service Level Agreement) tracking—proving to customers that your service met its promised availability. It also powers status pages, which show real-time or historical uptime data publicly.
TipMonitoring your own infrastructure prevents customers from discovering your downtime before you do. Set alerts at thresholds that matter: a single failed check may be noise, but three failures in a row usually signals a real problem.
The data from uptime monitoring—stored as uptime percentage over weeks or months—is essential for SLA compliance reporting and for identifying patterns in service degradation.