alerting rule
also: alert rule, monitoring rule, threshold rule
A configuration in monitoring systems that specifies conditions to trigger notifications or alerts when system metrics exceed thresholds or abnormal events occur.
An alerting rule is a conditional statement defined in monitoring tools like Prometheus, Grafana, or Nagios that evaluates metrics or logs and automatically sends notifications when criteria are met. Rules typically define a metric query, a threshold value, and a duration that the condition must persist before triggering.
For example, in Prometheus, an alerting rule might specify: "if CPU usage exceeds 80% for more than 5 minutes, send an alert to the operations team via email or Slack." This prevents noise from brief spikes while catching sustained problems that require attention.
Alerting rules are essential for proactive system management, allowing teams to respond to issues before users are affected. They can target various conditions: high memory consumption, disk space running low, service unavailability, or application error rates exceeding acceptable limits.