[feat][kubectl-plugin] add get workergroup cmd

We plan to add a [command to scale a worker group][1] in an existing RayCluster, e.g. `kubectl ray scale cluster (CLUSTER_NAME) (WORKER_GROUP)`. This requires the user to know the group name. There's currently no way to get the worker group names in a RayCluster other than getting the resource with kubectl and looking for or parsing out the names. A command to get worker groups details for a cluster might be helpful. [1]: ray-project#2926 Signed-off-by: David Xia <david@davidxia.com>
davidxia · Feb 12, 2025 · 99f6755 · 99f6755
1 parent 9559227
commit 99f6755
Show file tree

Hide file tree

Showing 18 changed files with 1,300 additions and 20 deletions.
diff --git a/kubectl-plugin/pkg/cmd/create/create_workergroup.go b/kubectl-plugin/pkg/cmd/create/create_workergroup.go
@@ -20,10 +20,6 @@ import (
 	"k8s.io/kubectl/pkg/util/templates"
 )
 
-const (
-	resourceNvidiaGPU = "nvidia.com/gpu"
-)
-
 type CreateWorkerGroupOptions struct {
 	configFlags       *genericclioptions.ConfigFlags
 	ioStreams         *genericclioptions.IOStreams
@@ -176,8 +172,8 @@ func createWorkerGroupSpec(options *CreateWorkerGroupOptions) rayv1.WorkerGroupS
 
 	gpuResource := resource.MustParse(options.workerGPU)
 	if !gpuResource.IsZero() {
-		podTemplate.Spec.Containers[0].Resources.Requests[corev1.ResourceName(resourceNvidiaGPU)] = gpuResource
-		podTemplate.Spec.Containers[0].Resources.Limits[corev1.ResourceName(resourceNvidiaGPU)] = gpuResource
+		podTemplate.Spec.Containers[0].Resources.Requests[corev1.ResourceName(util.ResourceNvidiaGPU)] = gpuResource
+		podTemplate.Spec.Containers[0].Resources.Limits[corev1.ResourceName(util.ResourceNvidiaGPU)] = gpuResource
 	}
 
 	return rayv1.WorkerGroupSpec{

diff --git a/kubectl-plugin/pkg/cmd/create/create_workergroup_test.go b/kubectl-plugin/pkg/cmd/create/create_workergroup_test.go
@@ -8,6 +8,8 @@ import (
 	"k8s.io/apimachinery/pkg/api/resource"
 	"k8s.io/utils/ptr"
 
+	"github.com/ray-project/kuberay/kubectl-plugin/pkg/util"
+
 	rayv1 "github.com/ray-project/kuberay/ray-operator/apis/ray/v1"
 )
 
@@ -34,14 +36,14 @@ func TestCreateWorkerGroupSpec(t *testing.T) {
 						Image: "DEADBEEF",
 						Resources: corev1.ResourceRequirements{
 							Requests: corev1.ResourceList{
-								corev1.ResourceCPU:    resource.MustParse("2"),
-								corev1.ResourceMemory: resource.MustParse("5Gi"),
-								resourceNvidiaGPU:     resource.MustParse("1"),
+								corev1.ResourceCPU:     resource.MustParse("2"),
+								corev1.ResourceMemory:  resource.MustParse("5Gi"),
+								util.ResourceNvidiaGPU: resource.MustParse("1"),
 							},
 							Limits: corev1.ResourceList{
-								corev1.ResourceCPU:    resource.MustParse("2"),
-								corev1.ResourceMemory: resource.MustParse("5Gi"),
-								resourceNvidiaGPU:     resource.MustParse("1"),
+								corev1.ResourceCPU:     resource.MustParse("2"),
+								corev1.ResourceMemory:  resource.MustParse("5Gi"),
+								util.ResourceNvidiaGPU: resource.MustParse("1"),
 							},
 						},
 					},

diff --git a/kubectl-plugin/pkg/cmd/get/get.go b/kubectl-plugin/pkg/cmd/get/get.go
@@ -24,5 +24,6 @@ func NewGetCommand(streams genericclioptions.IOStreams) *cobra.Command {
 	}
 
 	cmd.AddCommand(NewGetClusterCommand(streams))
+	cmd.AddCommand(NewGetWorkerGroupCommand(streams))
 	return cmd
 }
diff --git a/kubectl-plugin/pkg/cmd/get/get_cluster.go b/kubectl-plugin/pkg/cmd/get/get_cluster.go
@@ -40,6 +40,7 @@ func NewGetClusterCommand(streams genericclioptions.IOStreams) *cobra.Command {
 
 	cmd := &cobra.Command{
 		Use:               "cluster [NAME]",
+		Aliases:           []string{"clusters"},
 		Short:             "Get cluster information.",
 		SilenceUsage:      true,
 		ValidArgsFunction: completion.RayClusterCompletionFunc(cmdFactory),