Resource

dcloud_eks_nodegroup

A node group: compute instances of one server type that join a cluster. Its id is its name. With autoscaling = true the cluster autoscaler changes desired_size: add lifecycle { ignore_changes = [desired_size] } so an apply does not undo it.

Schema

Required

  • cluster_id (String) The id of the parent resource ({cluster} in the path). Changing it replaces the resource.
  • desired_size (Number) Nodes to run now, 0–100; with autoscaling on, the cluster autoscaler changes it between minSize and maxSize. From 0 to 100.
  • instance_type (String) An instance type, e.g. performance-4. The provider’s own type names are still accepted, deprecated. Up to 32 characters.
  • max_size (Number) Most nodes, 1–100; what the quotas count (nodes per node group, and nodes per cluster summed over its node groups; 20 and 50 by default): 409 QuotaExceeded above them. From 1 to 100.
  • min_size (Number) Fewest nodes, 0–100; the cluster autoscaler never goes below it. minSize ≤ desiredSize ≤ maxSize, else 400. From 0 to 100.
  • name (String) Node group name, unique in the cluster (409 when taken): lowercase letters, digits and hyphens, starting with a letter, at most 32 characters. Must match ^[a-z]([a-z0-9-]{0,30}[a-z0-9])?$. Changing it replaces the resource.

Optional

  • autoscaling (Boolean) Let the cluster autoscaler change desiredSize between minSize and maxSize (pending pods grow the group, underused nodes shrink it)
  • capacity_fallback (String) Stock fallback; default off. otherZones: zones outside zones may be used; equivalentTypes: an equivalent server type may be bought
  • drain_timeout_minutes (Number) Upgrades: minutes a node’s drain may wait on PodDisruptionBudgets; default 60. From 5 to 1440.
  • force_after_timeout (Boolean) Upgrades: after the timeout, remove the node anyway (true) or pause the upgrade (false, default)
  • instance_types (List of String) Instance types to buy, in order, when instanceType is sold out in a zone (same CPU architecture), e.g. performance-8
  • labels (Map of String) Kubernetes labels on every node, e.g. {"workload":"web"}; at most 20. kubernetes.io/, k8s.io/ and eks.ews.cloud/ keys are refused (400). Default none
  • node_role (String) Name of an IAM role the nodes act as; needs iam:PassRole on it. Up to 128 characters. Changing it replaces the resource.
  • tags (Map of String) Tags for the new resource, e.g. {"env": "prod", "project": "shop"}. Keys your organization marks required must be present, or the create is refused with the missing keys (GET /tags/api/keys lists them). Changing it replaces the resource.
  • taints (String) Kubernetes taints on every node: { key, value?, effect: NoSchedule | PreferNoSchedule | NoExecute }; at most 10. Default none
  • version (String) Kubernetes minor for the nodes; default the cluster’s, never above it One of: 1.36, 1.37.
  • zones (List of String) Availability zones the nodes spread over, balanced (e.g. ["de1a","de1b"]); default the region’s default zone. Applies to nodes created from then on

Read-Only

  • id (String) The resource's id.
  • status (String) creating, active, updating, degraded, deleting or failed.

Generated from the provider's schema.