An alarm is raised when a collected value falls outside its “normal” range of values. This “normal” range is defined by thresholds and severities configured for the alarm. A new alarm is raised each time the severity for the alarm changes. When the severity returns to normal, the alarm is canceled. Alarms critical to your operation can be configured to remain on view until they are acknowledged.
Displays for Spotlight alarms
Heat map | The Heat Map organizes connections based on the comparative number and severity of alarms raised against them. The Heat Map answers the question: Which monitored connections require the most urgent attention? |
Spotlight today | Spotlight Today is a display of recent alarms and alarms requiring acknowledgment. |
Alarms by time | Alarms by Time tracks the start time, duration, and severity of logged alarms. |
Alarm log | The alarm log is a display of logged alarms. |
Spotlight Overview page | The Spotlight Overview page illustrates the flows and components of a single monitored connection. The flows and components are updated in real time to highlight obvious bottlenecks and problem areas and color coded to indicate when an alarm is raised. |
Playback | Use the Playback facility to view events and data collected in the recent past on a Spotlight Overview Page, Drilldown or Spotlight Today. Playback time to capture the event or moment in Spotlight as though it was happening in real time. |
Act on (resolve, investigate) a raised alarm
From the Spotlight Overview page or Alarm log, right click on the component / alarm and select What’s this? to investigate the alarm by the following options.
Icon | Title | Description |
---|---|---|
Diagnose Connection drilldowns | Show performance statistics (grids and charts) related to the raised alarm. | |
Acknowledge | Acknowledge the alarm. This is appropriate for alarms that are configured to require acknowledgment. | |
Alarm History | Show the alarm in Alarms by Time. | |
Snooze | Temporarily remove the visual alert associated with the alarm. | |
Ignore | Configure Spotlight to ignore the alarm and future cases of the alarm. Select the rule by which you want Spotlight to ignore future cases: ignore this alarm for the current value or for the current connection. The choices are dependent on the type of alarm. This is a simplified interface for Do not alarm for certain values. | |
Alarm Settings | Configure the thresholds and severities that determine when the alarm is raised. |
Color
Spotlight uses color to identify the severity of an alarm. The alarm, Spotlight Overview page component or connection is colored. If more than one alarm is raised against the component or connection then the color shows the highest severity alarm raised.
Default Color | Severity |
---|---|
Disabled | |
Normal / No alarm is currently raised. | |
Information | |
Low | |
Medium | |
High |
Alternative colors for alarm severity are available. For more information, see Severity colors for accessibility.
Configuration
Many options are available from the Spotlight Client Configure ribbon. In particular:
Configure | Alarms | Set the thresholds and severities that determine when an alarm is raised. Disable an alarm. Set an alarm to require acknowledgment. Configure keyed alarms. Collect additional diagnostic information on an alarm. |
Configure | Alarm Actions | Set actions for Spotlight to take when an alarm is raised. The actions Spotlight can take include running a program and sending an email. Conditions on taking the action can be defined, such as the day of the week, the time of day, the severity of the alarm, the alarm type and the connection type. |
About each alarm
Information about each alarm is documented under the connection type.
- Analysis Services alarms
- Availability Group alarms
- Azure SQL Managed Instance alarms
- Hyper-V alarms
- SQL Azure alarms
- SQL Server Replication alarms
- SQL Server alarms
- Spotlight Diagnostic Server alarms
- Unix/Linux alarms
- VMware alarms
- Windows Server alarms