Machine Modifiers

Machine modifiers define configuration axes for machine deployments within test clusters. They are split into static modifiers (hardcoded) and dynamic modifiers (discovered at runtime from the infrastructure cluster).

Static Machine Modifiers

Group	Options	Description
`instance-type`	none, discovered instance types	KubeVirt VM instance type
`preference`	none, discovered preferences	KubeVirt VM preference
`dns-policy`	ClusterFirstWithHostNet, ClusterFirst, Default, None	Pod DNS resolution policy
`eviction-strategy`	LiveMigrate, External	VM eviction behavior
`network-multiqueue`	enabled, disabled	Network multi-queue support
`topology-spread-constraints`	enabled, disabled	Pod topology spread

Dynamic Machine Modifiers

These modifiers are discovered at runtime from the YAML configuration and infrastructure cluster:

Group	Source	Description
`cpu`	`resources.cpu` config values	VM CPU count
`memory`	`resources.memory` config values	VM memory size
`disk-size`	`resources.diskSize` config values	VM disk size
`storage-class`	StorageClasses from infra cluster	Storage backend
`node-affinity`	Node names from infra cluster	Node placement
`image-source`	`imageSources` config per distribution	OS disk image
`nameservers`	`nameservers` config	DNS nameservers

Datacenter Modifiers

Discovered from the KubeVirt infrastructure cluster:

Group	Source	Description
`vpc`	VPC CRDs (`kubeovn.io/v1`)	Virtual private cloud
`kubevirt-vpc-subnet`	Subnet CRDs per VPC	Network subnet
`kubevirt-storage-class`	StorageClasses from infra cluster	Storage class

How Modifiers Work

Each modifier follows the same pattern as cluster modifiers: a Name for human identification, a Group for exclusive selection, and a modify function for spec mutation.