Kubernetes 1.21: Finally, Stable Metrics for Reliable Monitoring!

Kubernetes 1.21 introduces the first officially supported stable metrics with the completion of the Metrics Stability Framework. These new stable metrics not only offer support guarantees but also provide safeguards when encountering problematic metrics.

You can view the list of stable Kubernetes metrics here.

What are stable metrics and why do we need them?

A stable metric is one that can be reliably consumed across multiple Kubernetes versions without the risk of ingestion errors. Metric stability is an ongoing concern for the community. The monitoring infrastructure of clusters often assumes the stability of certain control plane metrics. Therefore, we have introduced a mechanism for versioning metrics as a proper API, with stability guarantees around a formal process for deprecating metrics.

What are the stability levels for metrics?

Metrics can currently have one of two stability levels: alpha or stable.

Alpha metrics offer no stability guarantees; they can be changed or removed at any time. Currently, all Kubernetes metrics implicitly fall into this category.

Stable metrics can be guaranteed not to change unless the metric is marked as deprecated for a future Kubernetes version. By “not change,” we mean three things:

The metric itself will not be deleted or renamed.
The type of the metric will not change.
No labels can be added or removed from this metric.

From an ingestion perspective, it is backward compatible to add or remove possible values for already existing labels, but not the labels themselves. Therefore, it is allowed to add or remove values from an existing label. Stable metrics can also be marked as deprecated for a future Kubernetes version, as this is tracked in a metadata field and does not change the metric itself.

Adding or removing labels on stable metrics is not permitted. To add or remove a label from an existing stable metric, a new metric would need to be introduced and the stable one deprecated; otherwise, it would violate the compatibility agreements.

At ayedo, we support companies in efficiently utilizing Kubernetes and benefiting from the new developments. Take advantage of stable metrics for reliable monitoring of your Kubernetes clusters!

Source: Kubernetes Blog