-
Notifications
You must be signed in to change notification settings - Fork 706
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Adding cel validation on trainingRuntime CRD
Signed-off-by: Akshay Chitneni <achitneni@apple.com>
- Loading branch information
Akshay Chitneni
committed
Nov 11, 2024
1 parent
94dee0e
commit d023258
Showing
4 changed files
with
161 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
128 changes: 128 additions & 0 deletions
128
test/integration/controller.v2/trainingruntime_controller_test.go
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,128 @@ | ||
| /* | ||
| Copyright 2024 The Kubeflow Authors. | ||
| Licensed under the Apache License, Version 2.0 (the "License"); | ||
| you may not use this file except in compliance with the License. | ||
| You may obtain a copy of the License at | ||
| http://www.apache.org/licenses/LICENSE-2.0 | ||
| Unless required by applicable law or agreed to in writing, software | ||
| distributed under the License is distributed on an "AS IS" BASIS, | ||
| WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
| See the License for the specific language governing permissions and | ||
| limitations under the License. | ||
| */ | ||
|
|
||
| package controllerv2 | ||
|
|
||
| import ( | ||
| kubeflowv2 "github.com/kubeflow/training-operator/pkg/apis/kubeflow.org/v2alpha1" | ||
| testingutil "github.com/kubeflow/training-operator/pkg/util.v2/testing" | ||
| "github.com/kubeflow/training-operator/test/integration/framework" | ||
| "github.com/kubeflow/training-operator/test/util" | ||
| "github.com/onsi/ginkgo/v2" | ||
| "github.com/onsi/gomega" | ||
| corev1 "k8s.io/api/core/v1" | ||
| metav1 "k8s.io/apimachinery/pkg/apis/meta/v1" | ||
| "k8s.io/utils/ptr" | ||
| "sigs.k8s.io/controller-runtime/pkg/client" | ||
| ) | ||
|
|
||
| var _ = ginkgo.Describe("TrainingRuntime marker validations and defaulting", ginkgo.Ordered, func() { | ||
| var ns *corev1.Namespace | ||
|
|
||
| ginkgo.BeforeAll(func() { | ||
| fwk = &framework.Framework{} | ||
| cfg = fwk.Init() | ||
| ctx, k8sClient = fwk.RunManager(cfg) | ||
| }) | ||
| ginkgo.AfterAll(func() { | ||
| fwk.Teardown() | ||
| }) | ||
|
|
||
| ginkgo.BeforeEach(func() { | ||
| ns = &corev1.Namespace{ | ||
| TypeMeta: metav1.TypeMeta{ | ||
| APIVersion: corev1.SchemeGroupVersion.String(), | ||
| Kind: "Namespace", | ||
| }, | ||
| ObjectMeta: metav1.ObjectMeta{ | ||
| GenerateName: "training-runtime-marker-", | ||
| }, | ||
| } | ||
| gomega.Expect(k8sClient.Create(ctx, ns)).To(gomega.Succeed()) | ||
| }) | ||
| ginkgo.AfterEach(func() { | ||
| gomega.Expect(k8sClient.DeleteAllOf(ctx, &kubeflowv2.TrainingRuntime{}, client.InNamespace(ns.Name))).Should(gomega.Succeed()) | ||
| gomega.Expect(k8sClient.DeleteAllOf(ctx, &kubeflowv2.ClusterTrainingRuntime{})).Should(gomega.Succeed()) | ||
| }) | ||
|
|
||
| ginkgo.When("Creating TrainingRuntime", func() { | ||
| ginkgo.DescribeTable("Validate TrainingRuntime on creation", func(trainingRuntime func() *kubeflowv2.TrainingRuntime, errorMatcher gomega.OmegaMatcher) { | ||
| gomega.Expect(k8sClient.Create(ctx, trainingRuntime())).Should(errorMatcher) | ||
| }, | ||
| ginkgo.Entry("Should succeed to create trainingRuntime", | ||
| func() *kubeflowv2.TrainingRuntime { | ||
| return testingutil.MakeTrainingRuntimeWrapper(ns.Name, "runtime"). | ||
| Obj() | ||
| }, | ||
| gomega.Succeed()), | ||
| ginkgo.Entry("Should fail to create trainingRuntime with both MPI and Torch runtimes", | ||
| func() *kubeflowv2.TrainingRuntime { | ||
| runtime := testingutil.MakeTrainingRuntimeWrapper(ns.Name, "runtime").Obj() | ||
| runtime.Spec.MLPolicy = &kubeflowv2.MLPolicy{ | ||
| MLPolicySource: kubeflowv2.MLPolicySource{ | ||
| Torch: &kubeflowv2.TorchMLPolicySource{}, | ||
| MPI: &kubeflowv2.MPIMLPolicySource{}, | ||
| }, | ||
| } | ||
| return runtime | ||
| }, | ||
| testingutil.BeInvalidError()), | ||
| ginkgo.Entry("Should fail to create trainingRuntime with minNodes and torch.elasticPolicy", | ||
| func() *kubeflowv2.TrainingRuntime { | ||
| runtime := testingutil.MakeTrainingRuntimeWrapper(ns.Name, "runtime").Obj() | ||
| runtime.Spec.MLPolicy = &kubeflowv2.MLPolicy{ | ||
| NumNodes: ptr.To(int32(2)), | ||
| MLPolicySource: kubeflowv2.MLPolicySource{ | ||
| Torch: &kubeflowv2.TorchMLPolicySource{ | ||
| ElasticPolicy: &kubeflowv2.TorchElasticPolicy{}, | ||
| }, | ||
| }, | ||
| } | ||
| return runtime | ||
| }, | ||
| testingutil.BeInvalidError()), | ||
| ) | ||
| ginkgo.DescribeTable("Defaulting TrainingRuntime on creation", func(trainingRuntime func() *kubeflowv2.TrainingRuntime, wantTrainingRuntime func() *kubeflowv2.TrainingRuntime) { | ||
| created := trainingRuntime() | ||
| gomega.Expect(k8sClient.Create(ctx, created)).Should(gomega.Succeed()) | ||
| gomega.Expect(created).Should(gomega.BeComparableTo(wantTrainingRuntime(), util.IgnoreObjectMetadata)) | ||
| }, | ||
| ginkgo.Entry("Should succeed to default TorchMLPolicySource.NumProcPerNode=auto", | ||
| func() *kubeflowv2.TrainingRuntime { | ||
| runtime := testingutil.MakeTrainingRuntimeWrapper(ns.Name, "runtime").Obj() | ||
| runtime.Spec.MLPolicy = &kubeflowv2.MLPolicy{ | ||
| MLPolicySource: kubeflowv2.MLPolicySource{ | ||
| Torch: &kubeflowv2.TorchMLPolicySource{}, | ||
| }, | ||
| } | ||
| return runtime | ||
| }, | ||
| func() *kubeflowv2.TrainingRuntime { | ||
| runtime := testingutil.MakeTrainingRuntimeWrapper(ns.Name, "runtime").Obj() | ||
| runtime.Spec.MLPolicy = &kubeflowv2.MLPolicy{ | ||
| MLPolicySource: kubeflowv2.MLPolicySource{ | ||
| Torch: &kubeflowv2.TorchMLPolicySource{ | ||
| NumProcPerNode: ptr.To("auto"), | ||
| }, | ||
| }, | ||
| } | ||
| runtime.Spec.Template.Spec = testingutil.MakeJobSetWrapper(ns.Name, "runtime"). | ||
| Replicas(1).Obj().Spec | ||
| return runtime | ||
| }), | ||
| ) | ||
| }) | ||
| }) |