VPA resource policy configuration¶

VPA resource policy configuration refers to the settings within the resourcePolicy section of a VerticalPodAutoscaler (VPA) manifest that define constraints and control parameters for how the autoscaler manages container resources^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

Configuration Scope¶

The resource policy is configured under spec.resourcePolicy.containerPolicies^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md]. This list allows administrators to apply specific rules to different containers.

Target Selection¶

The containerName field determines which containers are affected by the policy^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

• Specific Container: You can target a specific container by its name.
• Wildcard (*): Using * specifies that the policy applies to all containers within the target Pod^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

Resource Constraints¶

The configuration allows for setting hard boundaries on the resources recommended by the VPA^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

minAllowed¶

This field sets the lower bound for resource requests^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md]. If the VPA's calculation suggests a value lower than this, the recommendation will adhere to this minimum.

maxAllowed¶

This field sets the upper bound for resource requests^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md]. It prevents the VPA from recommending resources that exceed this limit.

These boundaries influence the VPA's Lower Bound and Upper Bound recommendations seen in the status. If a Pod's current requests fall outside these bounds (or the calculated bounds), the VPA may trigger eviction (depending on the updateMode) to enforce the new settings^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

Resource Selection¶

The controlledResources field specifies which types of resources should be monitored and managed by the VPA^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].

• Supported Values: The available options are cpu and memory^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md].
• Usage: You can specify one or both types, allowing for fine-grained control over which resources are auto-scaled.

Example Configuration¶

The following YAML demonstrates a VPA configured with a resource policy that applies to all containers (*) in the target deployment^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md]:

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: hamster-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: hamster
  updatePolicy:
    updateMode: "Auto"
  resourcePolicy:
    containerPolicies:
      - containerName: '*'
        minAllowed:
          cpu: 100m
          memory: 50Mi
        maxAllowed:
          cpu: 1
          memory: 500Mi
        controlledResources: ["cpu", "memory"]

Related Concepts¶

• Vertical Pod Autoscaler
• Kubernetes

Sources¶

^[400-devops__06-Kubernetes__k8s-ithelp__Day27__README.md]