You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
"description": "Kueue provides the ability to admit workloads all in one (gang admission) and evicts workloads if they are not ready within a specific time.",
33247
33247
"type": "object",
33248
33248
"properties": {
@@ -33474,22 +33474,22 @@
33474
33474
"integrations"
33475
33475
],
33476
33476
"properties": {
33477
-
"gangSchedulingPolicy": {
33478
-
"description": "gangSchedulingPolicy controls how Kueue admits workloads. Gang Scheduling is the act of all or nothing scheduling, where workloads do not become ready within a certain period, they may be evicted and later retried. This field is optional.",
"description": "gangScheduling controls how Kueue admits workloads. Gang Scheduling is the act of all or nothing scheduling, where workloads do not become ready within a certain period, they may be evicted and later retried. This field is optional.",
"description": "integrations is a required field that configures the Kueue's workload integrations. Kueue has both standard integrations, known as job frameworks, and external integrations known as external frameworks. Kueue will only manage workloads that correspond to the specified integrations.",
"description": "preemption is the process of evicting one or more admitted Workloads to accommodate another Workload. Kueue has classical premption and preemption via fair sharing.",
33487
+
"description": "preemption is the process of evicting one or more admitted Workloads to accommodate another Workload. Kueue has classical premption and preemption via fair sharing. This field is optional.",
"description": "queueLabelPolicy controls how kueue manages workloads The default behavior of Kueue will manage workloads that have a queue-name label. Workloads that are missing these label will be ignored by Kueue. This field is optional.",
"description": "workloadManagement controls how kueue manages workloads. The default behavior of Kueue will manage workloads that have a queue-name label. Workloads that are missing these label will be ignored by Kueue. This field is optional.",
"description": "policy controls whether or not Kueue reconciles jobs that don't set the label kueue.x-k8s.io/queue-name. The allowed values are QueueNameRequired and QueueNameOptional. QueueNameOptional means that workloads will be suspended on creation and a label will be added via a mutating webhook. QueueNameRequired means that workloads that are managed by Kueue must have a label kueue.x-k8s.io/queue-name. If this label is not present on the workload, then Kueue will ignore this workload. Defaults to QueueNameRequired.",
"description": "RepositoryDigestMirrors holds cluster-wide information about how to handle mirros in the registries config. Note: the mirrors only work when pulling the images that are referenced by their digests.",
"description": "labelPolicy controls whether or not Kueue reconciles jobs that don't set the label kueue.x-k8s.io/queue-name. The allowed values are QueueName and None. None means that workloads will be suspended on creation and a label will be added via a mutating webhook. QueueName means that workloads that are managed by Kueue must have a label kueue.x-k8s.io/queue-name. If this label is not present on the workload, then Kueue will ignore this workload. Defaults to QueueName.",
// Kueue provides the ability to admit workloads all in one (gang admission)
210
211
// and evicts workloads if they are not ready within a specific time.
211
-
// +kubebuilder:validation:XValidation:rule="has(self.policy) && self.policy == 'ByWorkload' ? has(self.byWorkload) : !has(self.byWorkload)",message="byWorkload is required when policy is ByWorkload, and forbidden otherwise"
212
-
typeGangSchedulingPolicystruct {
212
+
// +kubebuilder:validation:XValidation:rule="has(self.policy) && self.policy == 'ByWorkload' ? has(self.byWorkload) : !has(self.byWorkload)",message="byWorkload is required when policy is byWorkload, and forbidden otherwise"
213
+
typeGangSchedulingstruct {
213
214
// policy allows you to enable and configure gang scheduling.
214
215
// This is an optional field.
215
216
// The allowed values are ByWorkload and Disabled.
@@ -218,7 +219,7 @@ type GangSchedulingPolicy struct {
218
219
// for admission as a single unit.
219
220
// Where workloads do not become ready over time, the entire workload may then be evicted and retried at a later time.
0 commit comments