Choose the trigger
Select the measured metric, comparison and threshold. Confirm the unit and the expected normal range for that device before activating the rule.
Alert engine
Give a threshold an operational meaning.
Create conditions against device metrics, then review the resulting events by severity and status. Rules express the conditions your team has decided require attention.
Control repeated events
Use the available cooldown setting to limit repeated alerts from the same condition. Review whether the chosen interval fits the speed at which the process changes.
Work the alert queue
Inspect the device and triggering reading, then acknowledge or resolve the event. Resolution notes give later reviewers context beyond the threshold value.