Alert types

Kloudfuse supports many alert types that support your observability experience.

Kloudfuse supports these specific alerting types:

Metric alerts

Monitor the metrics of continuous streams of data. The metrics that you collect through Kloudfuse or an API can trigger alerts when they cross a specified threshold.

Anomaly detection

Monitor anomalies to detect when a metric is behaving differently than it has in the past, accounting for trends and seasonal day-of-week and time-of-day patterns.

Type of Metric alert.

Change alerts

For each alert evaluation, Kloudfuse calculates the raw difference between the current time series the time series in the past (specified time interval), and proceed to compute the statistical aggregations (average, minimum, maximum, and sum) for the selected period. Kloudfuse triggers an alert when this computed series crosses the specified threshold.

Type of Metric alert.

Forecast alerts

Use forecast monitors to predict how a metric behaves in the future. When Kloudfuse issues alerts before a threshold is breached, you have the necessary time to proactively handle the underlying issues.

Type of Metric alert.

Outliers alerts

For each alert evaluation, Kloudfuse checks if all groups are clustered and exhibit similar behavior. It triggers an alert whenever at least one group diverges from the rest.

Type of Metric alert.

Threshold alerts

For each alert evaluation, Kloudfuse calculates the average, minimum, maximum, and sum aggregations over the selected period, and checks if they are above or below the threshold. Threshold alerts are a common practice when the business understands the range of expected values.

Type of Metric alert.

Log alerts

Log monitors generate alerts when a specified type of log exceeds a user-defined threshold over a period of time. Common use cases for log monitors include code exception errors and build job notifications.

APM alerts

While similar to standard metric alerts, we designed APM alerts to monitor features of APM metrics. With APM alerts, you can set alerts at the service level for metrics: number of hits, errors, and various latency indicators.

RUM alerts

Monitor the performance and user experience of real users, in real time. By analyzing metrics like page load time, JavaScript errors, and user interactions, you ensure a seamless client-side experience.

Traces alerts

Monitor traces to detect when a trace metric crosses a specified threshold over a particular period of time.

RUM alerts

Monitor events to detect when an event metric crosses a specified threshold over a particular period of time.