Dynatrace Operator release notes version 1.6.0

Latest Dynatrace
Release notes

Release date: June 30, 2025

Important notice

If you're running Dynatrace Operator version 1.6, we recommend upgrading to version 1.6.3 to receive the latest important patches.

Announcements

Dynatrace Operator version 1.6 introduces a new default and recommended DynaKube CRD version v1beta5. We encourage you to update your existing DynaKube resources to this latest version to take advantage of new features and enhancements.

Telemetry ingest endpoints

Enable Dynatrace telemetry endpoints in Kubernetes for cluster-local data ingest via DynaKube:

Easily ingest data via OTLP, Jaeger, StatsD or Zipkin endpoints
Analyze context-rich data with built-in apps, DQL, Notebooks and Dashboards

Read more in our about this new feature in the telemetry ingest guide.

FIPS-compliant Dynatrace Operator

Dynatrace Operator now provides FIPS-compliant images. For details, see our FIPS usage guide

New features and enhancements

OneAgent version 1.317+ The node image pull feature now supports the cloud-native full-stack mode.

Enrichment metadata is now propagated to pods based on the enrichment rules configured in Dynatrace. You can find more information about metadata enrichment in our user guide.

Dynatrace Operator now provides replacements for the external dependencies (binaries) used for the CSI liveness probe and the CSI node driver registrar. With mentioned replacements, increased frequency and promptness of updates will minimize vulnerabilities in the Dynatrace Operator.
- To use the new binaries, set the Helm switches csidriver.registrar.builtIn=true and csidriver.livenessprobe.builtIn=true.
- GKE Autopilot This configuration is not supported, since the CSI driver DaemonSet would no longer match the WorkloadAllowlists.
The old binaries will remain part of the operator image for future releases. The switch to new binaries by default will happen transparently with no action required and will be communicated in release notes. Until the final removal of binaries, vulnerability scanners might still report vulnerabilities. However, using the aforementioned Helm switches ensure that potentially vulnerable code will not be executed.

Dynatrace Operator components (Operator, Webhook, CSI driver) now comes with default sizeLimits for their ephemeral emptyDir volumes. Values are configurable via Helm to enable flexibility and guarantee reliable sizing.

It's now possible to reduce mounted paths for KSPM via the mappedHostPaths field in DynaKube. The DynaKube example on Github shows a recommended list of paths.

The Dynatrace Operator CSI driver's download job used for the node image pull feature has received new Helm switches for better configurability: csidriver.job.securityContext and csidriver.job.resources. Additionally, the existing Helm switches csidriver.annotations, csidriver.tolerations, and csidriver.labels are now also applied to the job definition.

A check has been added to prevent Helm upgrades if the DynaKube CRD is unavailable in the cluster to avoid installation or runtime errors.

Resolved issues

To ensure that OneAgents do not unnecessarily communicate with the in-cluster ActiveGate via the proxy, the IP address of the ActiveGate service is also passed to the NO_PROXY environment variable.

A new cluster role dynatrace-edgeconnect with a binding to the service account dynatrace-edgeconnect has been added. This allows EdgeConnect to run on OpenShift without the need of a user-defined SCC.

Improved OpenShift compatibility by avoiding hardcoded runAsUser values.

Changing from back-and-forth from CloudNative/HostMonitoring to ApplicationMonitoring will not cause problems for host CSI volume.

Known issues

We have identified the following known issues with Dynatrace Operator version 1.6.0–1.6.2.

In Kubernetes environments — especially those utilizing auto-scalers — there are challenges in reliably determining whether a node was intentionally removed or has failed unexpectedly. This ambiguity can lead to a high number of false-positive “Host is unavailable” alerts, impacting monitoring accuracy and alerting quality.

When switching from using the CSI driver without codeModulesImage to using it with node image pull, ensure that the CSI driver’s filesystem does not contain a code module for the specified DynaKube. If it does, the CSI driver will fail and require manual intervention to recover.

Removal and deprecation notices

The “Mark for Termination” event is deprecated and will be removed in a future Operator version. This functionality is now redundant, as it has been superseded by host availability events on host shutdown and reboot introduced in OneAgent version 1.301.

Upgrade from Dynatrace Operator version 1.5

There are no breaking changes in this version and all changes listed below are compatible:

The DynaKube field .spec.activeGate.persistentVolumeClaim has been renamed to .spec.activeGate.volumeClaimTemplate for clarity in DynaKube API version v1beta5 and later.