Skip to content

fix: add InPlace to VerticalPodAutoscaler v1 updateMode enum - #914

Open
starlightromero wants to merge 1 commit into
datreeio:mainfrom
starlightromero:fix/vpa-inplace-updatemode
Open

fix: add InPlace to VerticalPodAutoscaler v1 updateMode enum#914
starlightromero wants to merge 1 commit into
datreeio:mainfrom
starlightromero:fix/vpa-inplace-updatemode

Conversation

@starlightromero

Copy link
Copy Markdown
Contributor

Summary

Add InPlace as a valid updateMode value for the VerticalPodAutoscaler v1 schema (autoscaling.k8s.io/verticalpodautoscaler_v1.json).

Context

The InPlace update mode was introduced in VPA 1.7.0 as an alpha feature. It resizes pod resources (CPU/memory) without restart or eviction — if a resize cannot be applied, it defers and retries on the next reconciliation loop.

Currently, users with VPA 1.7.0+ using updateMode: InPlace get false validation failures from kubeconform when using this catalog as a schema source.

Change

 "enum": [
   "Off",
   "Initial",
   "Recreate",
   "InPlaceOrRecreate",
-  "Auto"
+  "Auto",
+  "InPlace"
 ]

References

Fixes #913

Add "InPlace" as a valid updateMode value for the
VerticalPodAutoscaler v1 schema. This mode was introduced in
VPA 1.7.0 as an alpha feature.

InPlace mode resizes pod resources without restart or eviction.
Unlike InPlaceOrRecreate, it never falls back to eviction — if
a resize cannot be applied it defers and retries.

Requires Kubernetes 1.33+ with InPlacePodVerticalScaling feature
gate and VPA --feature-gates=InPlace=true.

Ref: https://kubernetes.io/docs/concepts/workloads/autoscaling/vertical-pod-autoscale/#updateMode-InPlace
Fixes: datreeio#913
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Update VerticalPodAutoscaler v1 schema to include InPlace updateMode

1 participant