feat: Placement Group Support

Author: joshuakguo

charts/karpenter-crd/templates/karpenter.k8s.aws_ec2nodeclasses.yaml

@@ -516,6 +516,22 @@ spec:
                         - optional
                       type: string
                   type: object
+                placementGroupSelector:
+                  description: PlacementGroupSelector defines the name or the id of the placement to resolve with the nodeclass.
+                  properties:
+                    id:
+                      description: ID is the placement group id in EC2
+                      pattern: ^pg-[0-9a-z]+$
+                      type: string
+                    name:
+                      description: Name is the placement group name in EC2
+                      type: string
+                  type: object
+                  x-kubernetes-validations:
+                    - message: expected at least one, got none, ['name', 'id']
+                      rule: has(self.name) || has(self.id)
+                    - message: '''name'' and ''id'' are mutually exclusive'
+                      rule: '!(has(self.name) && has(self.id))'
                 role:
                   description: |-
                     Role is the AWS identity that nodes use.
@@ -814,6 +830,50 @@ spec:
                 instanceProfile:
                   description: InstanceProfile contains the resolved instance profile for the role
                   type: string
+                placementGroup:
+                  description: PlacementGroup contains the current placement group values that are available to this NodeClass.
+                  items:
+                    properties:
+                      id:
+                        description: The id for the placement group.
+                        pattern: ^pg-[0-9a-z]+$
+                        type: string
+                      name:
+                        description: The name for the placement group.
+                        type: string
+                      partitionCount:
+                        description: The partition count for the partition placement group.
+                        format: int32
+                        type: integer
+                      spreadLevel:
+                        description: The spread level for the spread placement group.
+                        enum:
+                          - host
+                          - rack
+                        type: string
+                      state:
+                        default: available
+                        description: The state of the placement group.
+                        enum:
+                          - available
+                          - pending
+                          - deleting
+                          - deleted
+                        type: string
+                      strategy:
+                        description: The strategy for the placement group.
+                        enum:
+                          - cluster
+                          - partition
+                          - spread
+                        type: string
+                    required:
+                      - id
+                      - name
+                      - state
+                      - strategy
+                    type: object
+                  type: array
                 securityGroups:
                   description: |-
                     SecurityGroups contains the current security group values that are available to the

charts/karpenter-crd/templates/karpenter.sh_nodeclaims.yaml

@@ -137,7 +137,7 @@ spec:
                           - message: label "kubernetes.io/hostname" is restricted
                             rule: self != "kubernetes.io/hostname"
                           - message: label domain "karpenter.k8s.aws" is restricted
-                            rule: self in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex"] || !self.find("^([^/]+)").endsWith("karpenter.k8s.aws")
+                            rule: self in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex", "karpenter.k8s.aws/placement-group-id", "karpenter.k8s.aws/placement-group-partition"] || !self.find("^([^/]+)").endsWith("karpenter.k8s.aws")
                       minValues:
                         description: |-
                           This field is ALPHA and can be dropped or replaced at any time

charts/karpenter-crd/templates/karpenter.sh_nodepools.yaml

@@ -225,7 +225,7 @@ spec:
                             - message: label "kubernetes.io/hostname" is restricted
                               rule: self.all(x, x != "kubernetes.io/hostname")
                             - message: label domain "karpenter.k8s.aws" is restricted
-                              rule: self.all(x, x in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex"] || !x.find("^([^/]+)").endsWith("karpenter.k8s.aws"))
+                              rule: self.all(x, x in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex", "karpenter.k8s.aws/placement-group-id", "karpenter.k8s.aws/placement-group-partition"] || !x.find("^([^/]+)").endsWith("karpenter.k8s.aws"))
                       type: object
                     spec:
                       description: |-
@@ -294,7 +294,7 @@ spec:
                                   - message: label "kubernetes.io/hostname" is restricted
                                     rule: self != "kubernetes.io/hostname"
                                   - message: label domain "karpenter.k8s.aws" is restricted
-                                    rule: self in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex"] || !self.find("^([^/]+)").endsWith("karpenter.k8s.aws")
+                                    rule: self in ["karpenter.k8s.aws/instance-tenancy", "karpenter.k8s.aws/capacity-reservation-type", "karpenter.k8s.aws/capacity-reservation-id", "karpenter.k8s.aws/capacity-reservation-interruptible", "karpenter.k8s.aws/ec2nodeclass", "karpenter.k8s.aws/instance-encryption-in-transit-supported", "karpenter.k8s.aws/instance-category", "karpenter.k8s.aws/instance-hypervisor", "karpenter.k8s.aws/instance-family", "karpenter.k8s.aws/instance-generation", "karpenter.k8s.aws/instance-local-nvme", "karpenter.k8s.aws/instance-size", "karpenter.k8s.aws/instance-cpu", "karpenter.k8s.aws/instance-cpu-manufacturer", "karpenter.k8s.aws/instance-cpu-sustained-clock-speed-mhz", "karpenter.k8s.aws/instance-memory", "karpenter.k8s.aws/instance-ebs-bandwidth", "karpenter.k8s.aws/instance-network-bandwidth", "karpenter.k8s.aws/instance-gpu-name", "karpenter.k8s.aws/instance-gpu-manufacturer", "karpenter.k8s.aws/instance-gpu-count", "karpenter.k8s.aws/instance-gpu-memory", "karpenter.k8s.aws/instance-accelerator-name", "karpenter.k8s.aws/instance-accelerator-manufacturer", "karpenter.k8s.aws/instance-accelerator-count", "karpenter.k8s.aws/instance-capability-flex", "karpenter.k8s.aws/placement-group-id", "karpenter.k8s.aws/placement-group-partition"] || !self.find("^([^/]+)").endsWith("karpenter.k8s.aws")
                               minValues:
                                 description: |-
                                   This field is ALPHA and can be dropped or replaced at any time

cmd/controller/main.go

@@ -17,6 +17,7 @@ package main
 import (
 	v1 "github.com/aws/karpenter-provider-aws/pkg/apis/v1"
 	"github.com/aws/karpenter-provider-aws/pkg/cloudprovider"
+	"github.com/aws/karpenter-provider-aws/pkg/cloudprovider/registrationhooks"
 	"github.com/aws/karpenter-provider-aws/pkg/controllers"
 	"github.com/aws/karpenter-provider-aws/pkg/operator"
 
@@ -60,6 +61,7 @@ func main() {
 			overlayUndecoratedCloudProvider,
 			clusterState,
 			op.InstanceTypeStore,
+			corecontrollers.WithRegistrationHook(registrationhooks.NewPlacementGroupRegistrationHook(op.GetClient(), op.InstanceProvider)),
 		)...).
 		WithControllers(ctx, controllers.NewControllers(
 			ctx,
@@ -84,6 +86,7 @@ func main() {
 			op.VersionProvider,
 			op.InstanceTypesProvider,
 			op.CapacityReservationProvider,
+			op.PlacementGroupProvider,
 			op.AMIResolver,
 		)...).
 		Start(ctx)