Thresholds are conditions that you can specify for selected parameters coming from managed OS 2200 and MCP systems. You can specify different thresholds for different systems, depending on the workload you expect on each system.
When you initially define a managed object, no threshold conditions are specified for it.
What Happens When a Threshold Condition Is Satisfied?
If a system parameter satisfies a threshold condition, Operations Sentinel Console raises an alert of major severity, which causes the following events to occur:
The Alerts icon under which the system appears assumes the shape and color indicating a major alert, unless an alert of higher severity is already outstanding.
A message describing the condition appears in the alerts pane.
The image that represents the system in its topology signifies a major alert, unless an alert of higher severity already exists for that system.
If the active alert policy includes raise actions for this threshold alert, those actions are executed.
The icon in the navigation pane that represents each zone containing the system assumes the shape and color indicating a major alert, unless an alert of higher severity already exists for a system in the zone.
What Happens When a Threshold Condition Ceases to Exist?
The alert raised for a threshold condition is automatically cleared once the threshold condition no longer exists. For example, suppose you consider it a serious situation when the number of demand runs on an OS 2200 system exceeds 200. You can configure Operations Sentinel to raise an alert when demand run 201 begins on that system. This warning signals you to this specific condition. When the number of demand runs decreases, the warning automatically disappears and the alert is cleared.
If the current alert policy has actions defined for this threshold alert
Any unexecuted raise or acknowledge actions are discarded.
Any actions associated with clearing of the threshold alert are executed.