Alerts

Using the Alert Threshold Wizard

The most straightforward way of adjusting alert thresholds is to use the alert threshold wizard. If you are adjusting threshold values at the instance or group levels you will always be using the wizard. If you are adjusting threshold values at the global level you have the option of either using the wizard or just entering the values directly. You can use the information below to help you either way, but it is presented in the context of the wizard.

Note: If you are looking to set a No Data alert for a datapoint (i.e. configure alerting in the absence of expected data), see the Datapoints section of the Creating a DataSource support article for details on how to accomplish this as it is not possible to set No Data alerts through the wizard.

The Threshold Wizard displays the level at which you are setting a threshold at the top, along with the relevant device/group, datasource, instance and datapoint information.  You can add thresholds that apply all day long, or during different time frames.  All thresholds will be displayed in the Effective Alert Timeline along the time period during which they are effective.  To add a new threshold, select the + icon:

1. Timeframe

Select a timeframe during which the threshold should be effective - you can either choose 'All Day' OR select a start time, which will automatically add another box for you to specify an end time. Note that time based thresholds are evaluated in order, so the threshold rows will order themselves by start time.

 

2. Comparison method

Select a comparison method - options include:

  • value - compare the datapoint value against a threshold
  • delta - compare the delta between the current datapoint and the previous datapoint against a threshold
  • NaNDelta - works the same as delta, but treats NaN values as 0
  • absolute value - compare the absolute value of the datapoint against a threshold
  • absolute delta - compare the absolute value of the delta between the current and last datapoints against a threshold
  • absolute NaNDelta - works the same as absolute delta, but treats NaN values as 0
  • absoluteDelta% - compare the absolute value of the percent change between the current datapoint and the previous datapoint against a threshold

 

3. Comparison operator

Select an operator - options include:

  • is equal to (=)
  • is not equal to (!=)
  • is greater than (>)
  • is less than (<)
  • greater than or equal to (>=)
  • less than or equal to (<=)

 

4. Specify values

Specify values for each alert severity level - you can specify up to three threshold values, one for triggering warning alerts, one for triggering error alerts and one for triggering critical alerts.

Skipping an Alert Level:

If you add two of the same threshold values for different severity levels, the lower severity level will be skipped. For example, if you set a warning threshold at 100 and an error threshold at 100, a datapoint value of 100 will trigger an error alert.

 

5. Edit global definition

Select this option to go to the datasource definition, from which you can set global thresholds that apply to all instances of the datasource.

 

Example 1

Consider the following alert threshold configuration:

Example 1

Between the hours of midnight and noon, if the percent change between two consecutive CPU values exceeds:

  • 30 a warning alert will be triggered
  • 40 an error alert will be triggered
  • 50 a critical alert will be triggered

Between the hours of noon and midnight, if the percent change between two consecutive BusyWorkers values exceeds:

  • 50 a warning alert will be triggered
  • 60 an error alert will be triggered
  • 70 a critical alert will be triggered

 

Example 2

Consider the following alert threshold configuration:

Example 2

All day long, if there is any change in the Status value, a warning alert will be triggered.