An item such as a database or a file server may encounter a condition that triggers an alert. An alert is a visual notification to the user that something might possibly go wrong with your server. By way of an example free space in your data files might be approaching zero or CPU usage on your web server might be close to 100%. Conditions that require the user's attention are reported as alerts. Alerts are shown on all views but not every alert is shown on every view and not every alerts may make sense for every item. For instance the data file free space alert does not make sense for a web server and is not triggered for one. All alerts are always reported on Enterprise view where you can see them by clicking on the alert icon next to the item the alert is being reported for.

See Also Enterprise View.

All alerts triggered for an Enterprise view item such as a database are saved for analysis and troubleshooting purposes. Alert history can be accessed by clicking on an Enterprise view item and selecting Alert history

See Also Alert.

All alerts currently active for an item can be accessed by clicking on a small alert icon in the right upper corner of the item they are triggered for. Alerts are ordered by severity first, then by the time (most recent first). If you right-click on an alert in the alert list you access a pop-up menu where you can choose to disable or configure the alert

See Also Alert.

If an alert is triggered for an item a small popup is displayed for a while next to this item. Alert notification stays on the screen for a few seconds to draw your attention to the item. After the notification balloon disappears you can still access alert list for the item by clicking on the alert icon next to it

See Also Alert.

The condition that triggers an alert can become more serious as the situation worsens or less serious if it improves. For example if there is less than 20% of free space on a drive it is worth drawing user's attention. If it becomes less that 10% then it is better be approached as soon as possible before the drive fills up. If it is approaching 0 it becomes critical. Correspondingly three severities can be defined per alert: informational, warning and critical in the order of importance depending on the amount of free space. You can change the default settings for an alert in the Alert Configuration dialog or turn off some severities altogether. For example if he chooses not to be notified unless the alert conditions become critical he can disable informational and warning severities for an alert

Some alerts can have only one severity defined at a time. Indeed it doesn't make sense to have multiple severities for a Server Down alert. It is critical, period. You can however change the severity level for such alert for example from critical to warning

See Also Alert.

The statistical data received from the servers can (and usually does) have short-term peaks. In most cases these peaks are irrelevant if the statistics is good over a longer-term period. For example a short-term spike in reads from disk may not be important if overall cache-hit ratio is 99%. To smoothen the data a technique called averaging is used. In its simplest form it means that the data is added up and distributed over a longer period so that lower-activity intervals would compensate for occasional spikes. You can choose whether to smoothen the data and if yes over what period

Main application view showing birds-eye overview of all items being monitored. As soon as an item is placed on the Enterprise view Insider starts collecting statistical data for the item and triggering alerts if necessary

An item on the Enterprise view. Can be a database, a file server, a web server, an application - anything you might be interested to see inside out.

A link between two or more items on the Enterprise view. Represents a relationship between items such as Oracle DataGuard primary-standby configuration

Detailed real-time snapshot of an Oracle database showing performance and other critical information for an Oracle instance

See Enterprise view item.

A link between two items on a view representing a relationships between items. For example a link between two Enterprise View items such as Oracle databases may represent a primary-standby relationship

See Also Enterprise link.

An item on some views showing the current number or the current amount and the maximum configured. For example a quantity gauge for Oracle user processes show current number of processes and the maximum defined by PROCESSES initialization parameter.

See Enterprise view item.

See Alert severity.

An item on some views showing the speed or intensity of a process. For example a speed gauge is used for the Activity column of the All Sessions table to show how active the session is i.e. how resource-intensive it is.

The data collected for an item for example an Oracle database. E.g for an Oracle database typically data dictionary views would be interrogated while for some other database it could be stored procedures or variables

A view is a visual image capturing certain aspects of real-time information for an item or items being monitored. For example Enterprise view is a birds-eye overview of all items while Instance view is a detailed picture for a specific Oracle database. In other words Instance view is a way of zooming in on an Oracle database item on the Enterprise view.

See Also Instance view, Enterprise View.