ManagedClusterPropertiesAutoScalerProfile Class
Definition
Important
Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.
Parameters to be applied to the cluster-autoscaler when enabled
public class ManagedClusterPropertiesAutoScalerProfiletype ManagedClusterPropertiesAutoScalerProfile = classPublic Class ManagedClusterPropertiesAutoScalerProfile- Inheritance
-
ManagedClusterPropertiesAutoScalerProfile
Constructors
| Name | Description |
|---|---|
| ManagedClusterPropertiesAutoScalerProfile() |
Initializes a new instance of the ManagedClusterPropertiesAutoScalerProfile class. |
| ManagedClusterPropertiesAutoScalerProfile(String, Nullable<Boolean>, Nullable<Boolean>, Nullable<Boolean>, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String) |
Initializes a new instance of the ManagedClusterPropertiesAutoScalerProfile class. |
| ManagedClusterPropertiesAutoScalerProfile(String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String, String) |
Initializes a new instance of the ManagedClusterPropertiesAutoScalerProfile class. |
Properties
| Name | Description |
|---|---|
| BalanceSimilarNodeGroups |
Gets or sets detects similar node pools and balances the number of nodes between them. Valid values are 'true' and 'false' |
| DaemonsetEvictionForEmptyNodes |
Gets or sets daemonSet pods will be gracefully terminated from empty nodes. If set to true, all daemonset pods on empty nodes will be evicted before deletion of the node. If the daemonset pod cannot be evicted another node will be chosen for scaling. If set to false, the node will be deleted without ensuring that daemonset pods are deleted or evicted. |
| DaemonsetEvictionForOccupiedNodes |
Gets or sets daemonSet pods will be gracefully terminated from non-empty nodes. If set to true, all daemonset pods on occupied nodes will be evicted before deletion of the node. If the daemonset pod cannot be evicted another node will be chosen for scaling. If set to false, the node will be deleted without ensuring that daemonset pods are deleted or evicted. |
| Expander |
Gets or sets the expander to use when scaling up. If not specified, the default is 'random'. See expanders for more information. Possible values include: 'least-waste', 'most-pods', 'priority', 'random' |
| IgnoreDaemonsetsUtilization |
Gets or sets should CA ignore DaemonSet pods when calculating resource utilization for scaling down. If set to true, the resources used by daemonset will be taken into account when making scaling down decisions. |
| MaxEmptyBulkDelete |
Gets or sets the maximum number of empty nodes that can be deleted at the same time. This must be a positive integer. The default is 10. |
| MaxGracefulTerminationSec |
Gets or sets the maximum number of seconds the cluster autoscaler waits for pod termination when trying to scale down a node. The default is 600. |
| MaxNodeProvisionTime |
Gets or sets the maximum time the autoscaler waits for a node to be provisioned. The default is '15m'. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| MaxTotalUnreadyPercentage |
Gets or sets the maximum percentage of unready nodes in the cluster. After this percentage is exceeded, cluster autoscaler halts operations. The default is 45. The maximum is 100 and the minimum is 0. |
| NewPodScaleUpDelay |
Gets or sets ignore unscheduled pods before they're a certain age. For scenarios like burst/batch scale where you don't want CA to act before the kubernetes scheduler could schedule all the pods, you can tell CA to ignore unscheduled pods before they're a certain age. The default is '0s'. Values must be an integer followed by a unit ('s' for seconds, 'm' for minutes, 'h' for hours, etc). |
| OkTotalUnreadyCount |
Gets or sets the number of allowed unready nodes, irrespective of max-total-unready-percentage. This must be an integer. The default is 3. |
| ScaleDownDelayAfterAdd |
Gets or sets how long after scale up that scale down evaluation resumes. The default is '10m'. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| ScaleDownDelayAfterDelete |
Gets or sets how long after node deletion that scale down evaluation resumes. The default is the scan-interval. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| ScaleDownDelayAfterFailure |
Gets or sets how long after scale down failure that scale down evaluation resumes. The default is '3m'. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| ScaleDownUnneededTime |
Gets or sets how long a node should be unneeded before it is eligible for scale down. The default is '10m'. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| ScaleDownUnreadyTime |
Gets or sets how long an unready node should be unneeded before it is eligible for scale down. The default is '20m'. Values must be an integer followed by an 'm'. No unit of time other than minutes (m) is supported. |
| ScaleDownUtilizationThreshold |
Gets or sets node utilization level, defined as sum of requested resources divided by capacity, below which a node can be considered for scale down. The default is '0.5'. |
| ScanInterval |
Gets or sets how often cluster is reevaluated for scale up or down. The default is '10'. Values must be an integer number of seconds. |
| SkipNodesWithLocalStorage |
Gets or sets if cluster autoscaler will skip deleting nodes with pods with local storage, for example, EmptyDir or HostPath. The default is true. |
| SkipNodesWithSystemPods |
Gets or sets if cluster autoscaler will skip deleting nodes with pods from kube-system (except for DaemonSet or mirror pods). The default is true. |