Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

PyTorchJob: Always show warnings when using elasticPolicy.nProcPerNode #2067

Merged
merged 1 commit into from
Apr 18, 2024
Merged

PyTorchJob: Always show warnings when using elasticPolicy.nProcPerNode #2067

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions pkg/webhooks/pytorch/pytorchjob_webhook.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,11 +88,13 @@ func validateSpec(spec trainingoperator.PyTorchJobSpec) (admission.Warnings, fie
var allErrs field.ErrorList
var warnings admission.Warnings

if spec.NprocPerNode != nil && spec.ElasticPolicy != nil && spec.ElasticPolicy.NProcPerNode != nil {
if spec.ElasticPolicy != nil && spec.ElasticPolicy.NProcPerNode != nil {
elasticNProcPerNodePath := specPath.Child("elasticPolicy").Child("nProcPerNode")
nprocPerNodePath := specPath.Child("nprocPerNode")
allErrs = append(allErrs, field.Forbidden(elasticNProcPerNodePath, fmt.Sprintf("must not be used with %s", nprocPerNodePath)))
warnings = append(warnings, fmt.Sprintf("%s is deprecated, use %s instead", elasticNProcPerNodePath.String(), nprocPerNodePath.String()))
if spec.NprocPerNode != nil {
allErrs = append(allErrs, field.Forbidden(elasticNProcPerNodePath, fmt.Sprintf("must not be used with %s", nprocPerNodePath)))
}
}
allErrs = append(allErrs, validatePyTorchReplicaSpecs(spec.PyTorchReplicaSpecs)...)
return warnings, allErrs
Expand Down
31 changes: 31 additions & 0 deletions pkg/webhooks/pytorch/pytorchjob_webhook_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -218,6 +218,37 @@ func TestValidateV1PyTorchJob(t *testing.T) {
field.Forbidden(pytorchReplicaSpecPath.Key(string(trainingoperator.PyTorchJobReplicaTypeMaster)).Child("replicas"), ""),
},
},
"Spec.ElasticPolicy.NProcPerNode are set": {
pytorchJob: &trainingoperator.PyTorchJob{
ObjectMeta: metav1.ObjectMeta{
Name: "test",
},
Spec: trainingoperator.PyTorchJobSpec{
ElasticPolicy: &trainingoperator.ElasticPolicy{
NProcPerNode: ptr.To[int32](1),
},
PyTorchReplicaSpecs: map[trainingoperator.ReplicaType]*trainingoperator.ReplicaSpec{
trainingoperator.PyTorchJobReplicaTypeMaster: {
Replicas: ptr.To[int32](1),
Template: corev1.PodTemplateSpec{
Spec: corev1.PodSpec{
Containers: []corev1.Container{
{
Name: "pytorch",
Image: "gcr.io/kubeflow-ci/pytorch-dist-mnist_test:1.0",
},
},
},
},
},
},
},
},
wantWarnings: admission.Warnings{
fmt.Sprintf("%s is deprecated, use %s instead",
specPath.Child("elasticPolicy").Child("nProcPerNode"), specPath.Child("nprocPerNode")),
},
},
"Spec.NprocPerNode and Spec.ElasticPolicy.NProcPerNode are set": {
pytorchJob: &trainingoperator.PyTorchJob{
ObjectMeta: metav1.ObjectMeta{
Expand Down
Loading