| page_title | cozystack_scheduling_class Resource - cozystack |
|---|---|
| subcategory | |
| description | A Cozystack SchedulingClass (cluster-scoped, cozystack.io group): named placement policy (nodeSelector/affinities/topologySpreadConstraints). The spec is surfaced as JSON. |
A Cozystack SchedulingClass (cluster-scoped, cozystack.io group): named placement policy (nodeSelector/affinities/topologySpreadConstraints). The spec is surfaced as JSON.
resource "cozystack_scheduling_class" "gpu" {
name = "gpu"
spec = jsonencode({
nodeSelector = { "nvidia.com/gpu.present" = "true" }
})
}name(String) SchedulingClass name (metadata.name). Immutable.
spec(String) Full SchedulingClass spec as JSON (nodeSelector,nodeAffinity,podAffinity,podAntiAffinity,topologySpreadConstraints).wait_for_ready(Boolean) Block on create/update until the tenant'sReadycondition is true.wait_timeout(String) Maximum time to wait whenwait_for_readyis set (Go duration, e.g.10m).
chart_version(String) Deployed chart version (status.version).id(String) Object name (cluster-scoped).ready(Boolean) Whether the application'sReadycondition is true.uid(String) Server-assigned object UID (metadata.uid). Stable across updates; changes on recreate.
Import is supported using the following syntax:
The terraform import command can be used, for example:
# Cluster-scoped: import by name only.
terraform import cozystack_scheduling_class.example name