Operator now has Long-Term Support (LTS) version

VictoriaMetrics Operator has been developing at a neck-breaking pace, bringing numerous improvements, features, and fixes to our community. We usually make at least a single release every two weeks. While this rapid iteration cycle is great for delivering fixes and improvements quickly, it can be challenging for administrators managing critical production environments.

Frequent updates mean that changes in the operator may sometimes cause unexpected pod rollouts, restarts, or configuration modifications. For large-scale deployments, managing these frequent updates requires significant time and effort to test. To address these concerns and provide a more predictable, stable path for our enterprise and production users, we are would like to introduce Long-Term Support (LTS) releases for the VictoriaMetrics Operator.

The v0.68 release is the current LTS version and brings many stability improvements and features, including:

• Introduction of the VMDistributed Custom Resource (CR) to easily deploy and manage VictoriaMetrics Distributed cluster components.
• Vertical Pod Autoscaler (VPA) support for vmcluster, vtcluster, vlcluster and vmauth.
• The statefulRollingUpdateStrategyBehavior property for managing VMAgent update strategy in a stateful mode.
• Improved logging with prettified reconcile diffs showing only changed JSON fields.

See latest v0.68.x release notes here and LTS-specific changelog is available here.

LTS releases include not only bug fixes and security patches, but also new features and improvements that are considered stable and suitable for production use. An example of such feature is dry-run mode.

One of the features that help with operator maintenance is the ability to see what changes will be applied without actually making them. Changes in the operator may sometimes be destructive or unexpected, so it is important to have a mode where diffs are printed, but not applied.

With the new dry-run mode, you can run the operator using the --dry-run argument or the DRY_RUN environment variable. In this mode, the operator uses a fake client that logs API calls instead of executing them against the Kubernetes API server. These calls also have a diff to be applied printed first. This is especially useful during operator upgrades to ensure that the upgrade won’t cause any unexpected mutating changes.

We are committed to supporting every LTS release for a year and marking one of our releases as LTS every 6 months.

This allows our users, who prefer stability over new features, to stay on LTS releases. There is a 6 months overlap between LTS releases, which gives enough time for the upgrade to the next LTS release.