PodGroupSpec
io.k8s.api.scheduling.v1alpha3.PodGroupSpec
source ↗
PodGroupSpec defines the desired state of a PodGroup.
disruptionModeDisruptionModedisruptionMode defines the mode in which a given PodGroup can be disrupted. Controllers are expected to fill this field by copying it from a PodGroupTemplate. One of Single, All. Defaults to Single if unset. This field is immutable.
allAllDisruptionModeall specifies that all children can only be disrupted together.singleSingleDisruptionModesingle specifies that children can be disrupted independently from each other.parentCompositePodGroupNamestringparentCompositePodGroupName contains the name of the parent composite pod group within the same namespace as this pod group. If it's nil, then this pod group is a root of a workload's hierarchy. This field is used only when the CompositePodGroup feature gate is enabled. This field is immutable.preemptionPolicystringpreemptionPolicy is the Policy for preempting pods/podgroups with lower priority. One of Never, PreemptLowerPriority. Defaults to PreemptLowerPriority if unset. When Priority Admission Controller is enabled, it populates this field from PriorityClassName, and defaults to PreemptLowerPriority if value is unset in PriorityClass. This field is immutable. This field is available only when the PodGroupPreemptionPolicy feature gate is enabled.priorityinteger (int32)priority is the value of priority of this pod group. Various system components use this field to find the priority of the pod group. When Priority Admission Controller is enabled, it prevents users from setting this field. The admission controller populates this field from PriorityClassName. The higher the value, the higher the priority. This field is immutable.priorityClassNamestringpriorityClassName defines the priority that should be considered when scheduling this pod group. Controllers are expected to fill this field by copying it from a PodGroupTemplate. Otherwise, it is validated and resolved similarly to the PriorityClassName on PodGroupTemplate (i.e. if no priority class is specified, admission control can set this to the global default priority class if it exists. Otherwise, the pod group's priority will be zero). This field is immutable.resourceClaims[]PodGroupResourceClaimresourceClaims defines which ResourceClaims may be shared among Pods in the group. Pods consume the devices allocated to a PodGroup's claim by defining a claim in its own Spec.ResourceClaims that matches the PodGroup's claim exactly. The claim must have the same name and refer to the same ResourceClaim or ResourceClaimTemplate.This is a beta-level field and requires that the DRAWorkloadResourceClaims feature gate is enabled.This field is immutable.
namestring requiredname uniquely identifies this resource claim inside the PodGroup. This must be a DNS_LABEL.resourceClaimNamestringresourceClaimName is the name of a ResourceClaim object in the same namespace as this PodGroup. The ResourceClaim will be reserved for the PodGroup instead of its individual pods.Exactly one of ResourceClaimName and ResourceClaimTemplateName must be set.resourceClaimTemplateNamestringresourceClaimTemplateName is the name of a ResourceClaimTemplate object in the same namespace as this PodGroup.The template will be used to create a new ResourceClaim, which will be bound to this PodGroup. When this PodGroup is deleted, the ResourceClaim will also be deleted. The PodGroup name and resource name, along with a generated component, will be used to form a unique name for the ResourceClaim, which will be recorded in podgroup.status.resourceClaimStatuses.This field is immutable and no changes will be made to the corresponding ResourceClaim by the control plane after creating the ResourceClaim.Exactly one of ResourceClaimName and ResourceClaimTemplateName must be set.schedulingConstraintsPodGroupSchedulingConstraintsschedulingConstraints defines optional scheduling constraints (e.g. topology) for this PodGroup. Controllers are expected to fill this field by copying it from a PodGroupTemplate. This field is immutable. This field is only available when the TopologyAwareWorkloadScheduling feature gate is enabled.
topology[]TopologyConstrainttopology defines the topology constraints for the pod group. Currently only a single topology constraint can be specified. This may change in the future.
keystring requiredkey specifies the key of the node label representing the topology domain. All pods within the PodGroup must be colocated within the same domain instance. Different PodGroups can land on different domain instances even if they derive from the same PodGroupTemplate. Examples: "topology.kubernetes.io/rack"schedulingPolicyPodGroupSchedulingPolicy requiredschedulingPolicy defines the scheduling policy for this instance of the PodGroup. Controllers are expected to fill this field by copying it from a PodGroupTemplate.
basicBasicSchedulingPolicybasic specifies that the pods in this group should be scheduled using standard Kubernetes scheduling behavior. Setting this field at group creation time opts this group to basic scheduling; this field cannot be changed afterward.gangGangSchedulingPolicygang specifies that the pods in this group should be scheduled using all-or-nothing semantics. Setting this field at group creation time opts this group to gang scheduling; this field cannot be set or unset afterward. The minCount field within Gang scheduling policy remains mutable after group creation.
minCountinteger (int32) requiredminCount is the minimum number of pods that must be schedulable or scheduled at the same time for the scheduler to admit the entire group. It must be a positive integer. This field is mutable to support workload scaling.Note that the scheduler operates on an eventually consistent model. Updates to minCount may not be immediately reflected in scheduling decisions due to propagation delays. If minCount is updated while a scheduling cycle is in progress for that group, the new value may not take effect until the next cycle. Moreover, minCount is only enforced during scheduling, meaning that modifications to this field do not affect already-scheduled pods, applying only to those evaluated in future cycles.workloadRefWorkloadReferenceworkloadRef references an optional PodGroup template within the Workload object that was used to create the PodGroup. This field is immutable.
templateNamestring requiredtemplateName is the name of a template within the Workload object that was used to create a pod group. It must be a DNS label. This field is required.workloadNamestring requiredworkloadName is the name of the Workload object that contains a template that was used when creating a pod group. It must be a DNS name. This field is required.