Upgrade Guide - Metric Alerting

How-to guide
6-min read
Published Jun 15, 2025

In Dynatrace classic metric alerting, you use custom metric alerting with thresholds and baselines along with built-in metric anomaly detectors. This document serves as a guide for upgrading to metric alerting in the latest Dynatrace, ensuring that scenarios that raise alerts on metric thresholds and baselines continue working seamlessly.

New concepts

Before you start with performing this upgrade, familiarize yourself with core concepts of the latest Dynatrace, such as what Dynatrace Grail is and how to query data using DQL.

Anomaly detection

Anomaly detection is the automated process of identifying unusual patterns or behaviors within data that deviate from the norm. This involves continuous monitoring of application performance, user actions, and various metrics to establish baseline metrics and detect any deviations in real-time. Customizable thresholds and automated baselining help in accurately detecting and notifying anomalies, ensuring prompt alerts for potential issues.

In the latest Dynatrace, anomaly detectors help with the following:

Getting notified about problems in an environment.
Using Davis analyzer (static and auto-adoptive thresholds, or seasonal baseline) to detect abnormal behavior and set up alerts for custom events.

Prerequisites

An active DPS license
Davis Anomaly Detection permissions

Metric alerting in Dynatrace Classic

The main use case for configuring and applying thresholds and baselines on metrics and timeseries in general is automatic detection of outliers and anomalies and raising alerts for those.

Dynatrace by default ships a large list of out-of-the-box anomaly detectors. Extensions and users have the possibility to introduce their own completely customized anomaly detectors.

Out-of-the-box detectors in the previous Dynatrace

Host and process infrastructure anomaly detectors

Detecting CPU and memory saturation along with network, host, process, and process groups instance availability detection. For more information, see Adjust the sensitivity of anomaly detection for infrastructure.

Automated multi-dimensional baselining detector

Automated baselining attempts to determine the best reference values for incoming application and service traffic. For more information, see the following topics:

Extension events

Installed extensions providing various event thresholds that are evaluated by anomaly detection. For more information, see Adjust the sensitivity of anomaly detection for extension events.

Customizable classic anomaly detectors in the previous Dynatrace

Metric events

Dynatrace offers customizable anomaly detectors for domain specific business needs. Once an anomaly detector is triggered, an event is created based on metric data.

There are two types of metric events based on how the metric is queried for event evaluation:

Metric key events evaluate the incoming measures of a single metric. You can use only static thresholds with this query type.
Metric selector events evaluate a complex query defined by the metric selector. This query type can include historical data and even arithmetic operations with multiple metrics.

To learn more about the differences between metric key and metric selector and how to set them up in classic settings, see Dynatrace Metric Events – Setup anomaly detection based on your business.

Metric events used by other solutions

AWS

Dynatrace offers recommended AWS custom alerting rules. Enabling predefined alerting rules will create such configurations in metric events. For more information, see Set up metric events for alerting.

Azure

Dynatrace offers recommended Azure custom alerting rules. Enabling custom alerting rules will create such configurations in metric events. For more information, see Set up metric events for alerting.

SLO

In Service-Level Objectives (previous Dynatrace) or SLOs Service-Level Objectives Classic you have the option to create a burn rate or status alerts. For more information, see Configure and monitor service-level objectives with Dynatrace.

SLO use functional metrics which behave different during transformation. We're working on a separated guide, covering SLOs and how to upgrade them.

Extensions

For hundreds of technologies, extensions make domain-specific expertise available in Dynatrace by adding individual data acquisition capabilities. Some extensions—such as the one for F5—provide predefined alerts that are automatically created as metric events during installation. For more information, see the F5 BIG-IP extension in Dynatrace Hub.

F5 Big IP extension in settings

You can use the transpiler from Davis Anomaly Detection to update the respective configuration.

Once you're done, verify whether each configuration was successfully upgraded and whetherany warnings or errors occurred during the upgrade. In the future, there'll be a DQL Alerting template similar to the one available for F5 BIG-IP.

How to upgrade metric events to anomaly detectors

To create an anomaly detector, use Davis Anomaly Detection.

Davis Anomaly Detection

Davis Anomaly Detection - new Davis Anomaly Detection focuses on enhancing timeseries monitoring through real-time analysis of data stored in Grail using DQL queries. We encourage you to also try the app for more advanced configuration options such as:

Using DQL queries in addition to Grail records.
Alerting on data such as logs, spans, and business events.
Creating advanced queries and including a higher number of data records.

Before continuing, have a look at the following resources for more detailed information:

Example: Upgrade of a metric event from Dynatrace Classic

Enable authorization settings

If you're running Davis Anomaly Detection - new Davis Anomaly Detection for the first time, enable authorization settings.

Go to Davis Anomaly Detection.
Select > Authorization settings.
Select the required permissions under Permissions.

Choose metrics for upgrading

Go to Davis Anomaly Detection.
Select Anomaly detector > Improve metric events with DQL to fetch all available metric events. Only metric selectors are available to transform.
Select the metric events you want to transform and select Transform.

You can only transform metric selectors.

Transformation

The metric event we selected for transformations is automatically disabled, and the newly created configuration is active instead.

If everything is fine, a successful notification appears in the lower-right corner of the screen.

Verify that everything works as expected

After the transformation is done, the app checks whether the created query meets the requirements of the configuration. However, this is not guaranteed, as, for example, the check can't detect whether the data from the metric is available or the necessary tags are present.

You therefore need to manually check after the transformation:

Go to Davis Anomaly Detection.
Select > Open with and select View and execute Grail query without leaving the app to check whether the metric is working as expected.

Congratulations! You have just transformed your first Dynatrace classic metric event!

The next time you access the Transformation page in Davis Anomaly Detection - new Davis Anomaly Detection, you'll see that the transformed metric event has changed its State to Disabled, and its Migrated status to Accept Yes.

Davis Anomaly Detection app