Adding Trigger Rules to an Alarm

Define one or more trigger rules, or predicates, for an alarm. A trigger rule is a condition (defined by the query) that must be satisfied for the alarm to be in the firing state, and also includes severity, trigger delay (pendingDuration), and the alarm body to include in notifications. A condition in a trigger rule can specify a threshold, such as 90% for CPU utilization, or an absence.

Add up to two trigger rules, or predicates, to an alarm. For example, add a critical trigger rule for 95 percent and a warning trigger rule for 90 percent.

Note

To understand how notifications are sent when an alarm has multiple trigger rules, see Grouping Notifications for an Alarm and Splitting Notifications for an Alarm.

Example trigger rules in an alarm

{
  "body": "95% CPU utilization",
  "pendingDuration": "PT5M"
  "query": "CPUUtilization[1m].mean()>95",
  "ruleName": "95",
  "severity": "CRITICAL",
},
{
  "body": "90% CPU utilization",
  "pendingDuration": "PT5M"
  "query": "CPUUtilization[1m].mean()>90",
  "ruleName": "90",
  "severity": "WARNING",
}

For valid predicate operators in MQL expressions, see Predicate Operators.

For alarm troubleshooting, see Troubleshooting Alarms.

1. Open the navigation menu and click Observability & Management. Under Monitoring, click Alarm Definitions.
2. Click the name of the alarm that you want to update.
3. On the alarm details page, click Actions and then select Edit alarm.
4. In the Metric description area, verify that the metric namespace and the metric name that you want are selected.
5. To define a trigger rule (predicate) by using Basic mode (default), go to the Trigger rule area and provide the following values:
  
  For Operator, select the operator to use in the condition threshold. See Predicate Operators.
  
  For Value, enter the value to use for the condition threshold. For the between and outside operators, enter both values for the range.
  
  For Trigger delay minutes, enter the number of minutes that the condition must be maintained before the alarm is in the firing state.
  
  For Alarm severity, select the perceived type of response required when the alarm is in the firing state for this condition (trigger rule).
  
  For Alarm body, enter the human-readable content of the notification for this condition (trigger rule).
  We recommend providing guidance to operators for resolving the alarm condition. Consider adding links to standard runbook practices. Example: "High CPU usage alert. Follow runbook instructions for resolution."
  
  To add another trigger rule (condition), click Additional trigger rule.
6. To define a trigger rule (predicate) by updating the MQL expression, perform the following steps:
  
  At the top of the Edit alarm page, click Switch to Advanced Mode.
  
  Provide the following values:
  
  For Trigger delay minutes, enter the number of minutes that the condition must be maintained before the alarm is in the firing state.
  
  For Alarm severity, select the perceived type of response required when the alarm is in the firing state for this condition (trigger rule).
  
  For Query code editor, edit the MQL query to specify the trigger rule that you want.
  The trigger rule is the operator and value fragment in the MQL query. For example, the fragment > 90 represents the greater-than operator and a 90 value. Example MQL query:
  CpuUtilization[5m].Mean() > 90
  For reference, see Querying Metric Data and Monitoring Query Language (MQL) Reference.
  
  For Alarm body, enter the human-readable content of the notification for this condition (trigger rule).
  
  To add another trigger rule (condition), click Additional trigger rule.
7. Update any other values for the alarm, as needed.
  For more information about the fields, see Creating a Basic Alarm.
8. Click Save alarm.
Use the oci monitoring alarm update command and required parameters to update trigger rules in an alarm. A trigger rule is part of the MQL expression, or query.

For an alarm with one rule, use the --query-text parameter. Example updating trigger rule to greater than 90 percent (>90):
oci monitoring alarm update --query-text "CPUUtilization[1m].mean()>90" [...]
For an alarm with multiple rules, use the --overrides parameter. Example:
oci monitoring alarm update --overrides <json_file_path> [...]
Example JSON file for request (alarm with multiple trigger rules)
{ "body": "95% CPU utilization", "pendingDuration": "PT5M" "query": "CPUUtilization[1m].mean()>95", "ruleName": "95", "severity": "CRITICAL", }, { "body": "90% CPU utilization", "pendingDuration": "PT5M" "query": "CPUUtilization[1m].mean()>90", "ruleName": "90", "severity": "WARNING", }
For a complete list of parameters and values for CLI commands, see the Command Line Reference for Monitoring.
Run the UpdateAlarm operation to update an alarm. A trigger rule is part of the MQL expression in a query attribute.

Provide the trigger rules using UpdateAlarmDetails. For an alarm with one rule (no overrides value), use the query attribute. For an alarm with multiple rules, use the overrides attribute.

Oracle Cloud Infrastructure Documentation

Adding Trigger Rules to an Alarm