KEP-2170: Add TrainJob conditions #2322

tenzen-y · 2024-11-07T14:47:03Z

What this PR does / why we need it:
I implemented the TrainJob condition mechanism based on https://github.com/kubeflow/training-operator/tree/master/docs/proposals/2170-kubeflow-training-v2#state-transition

However, the current implementation depends on the JobSet status.conditions as opposed to the status.terminalState since the terminalState was introduced in JobSet v0.6, then the JobSet depends on the K8s lib. After we upgrade the training-operator dep version to 1.30 in #2299, we can rely on the termonalState.

So, after we upgrade the K8s libs to 1.30, we can revisit the JobSet status.terminalState.

Which issue(s) this PR fixes (optional, in Fixes #<issue number>, #<issue number>, ... format, will close the issue(s) when PR gets merged):
Part-of: #2207
Relates to #2170

Checklist:

Docs included if any changes are user facing

tenzen-y · 2024-11-07T20:49:05Z

test/integration/framework/framework.go

@@ -51,7 +51,7 @@ type Framework struct {
 }

 func (f *Framework) Init() *rest.Config {
-	log.SetLogger(zap.New(zap.WriteTo(ginkgo.GinkgoWriter), zap.UseDevMode(true)))
+	ctrl.SetLogger(zap.New(zap.WriteTo(ginkgo.GinkgoWriter), zap.Level(zapcore.Level(-5)), zap.UseDevMode(true)))


Surfacing the debug logs.

coveralls · 2024-11-07T20:50:18Z

Pull Request Test Coverage Report for Build 11754225694

Details

1 of 1 (100.0%) changed or added relevant line in 1 file are covered.
No unchanged relevant lines lost coverage.
Overall coverage remained the same at 100.0%

Totals
Change from base Build 11663764609:	0.0%
Covered Lines:	77
Relevant Lines:	77

💛 - Coveralls

tenzen-y · 2024-11-07T20:52:52Z

/hold for review

tenzen-y · 2024-11-07T20:57:40Z

/assign @kubeflow/wg-training-leads

Signed-off-by: Yuki Iwai <[email protected]>

andreyvelich

Thank you for this @tenzen-y!
I left my comments.
/assign @kubeflow/wg-training-leads @varshaprasad96 @akshaychitneni @deepanker13 @helenxie-bit @Electronic-Waste @saileshd1402 @kannon92 @kuizhiqing @shravan-achar @seanlaii

andreyvelich · 2024-11-08T15:17:42Z

pkg/apis/kubeflow.org/v2alpha1/trainjob_types.go

@@ -48,6 +48,43 @@ type TrainJob struct {
 	Status TrainJobStatus `json:"status,omitempty"`
 }

+const (
+	// TrainJobSuspended means the TrainJob is suspended.


Suggested change

// TrainJobSuspended means the TrainJob is suspended.

// TrainJobSuspended means that the TrainJob is suspended.

andreyvelich · 2024-11-08T15:18:27Z

pkg/apis/kubeflow.org/v2alpha1/trainjob_types.go

+	TrainJobSuspended string = "Suspended"
+
+	// TrainJobComplete means that the TrainJob has completed its execution.
+	TrainJobComplete string = "Complete"


Please can you create an issue in JobSet to rename Completed condition to Complete as we discussed?

Yeah, we can

andreyvelich · 2024-11-08T15:22:13Z

pkg/apis/kubeflow.org/v2alpha1/trainjob_types.go

+	// TrainJobJobsBuildFailedReason is the "Created" condition reason.
+	// When the building objects based on the TrainJob and the specified runtime failed,
+	// this is added.
+	TrainJobJobsBuildFailedReason string = "JobsBuildFailed"


Don't we need this reason as well: JobsBuildSucceeded ?
E.g. transition stage in which objects have been built, but not yet deployed ?

In that case, the Created reason isJobsCreationFailed.

Basically, we preserve the failed reason rather than succeeded reason (e.g., JobsBuildSucceeded vs JobsCreationFailed)

You are right, so the reasons should be the one that user can always see on the TrainJob.
E.g. JobsBuildSucceeded is the transition reason which will be always overridden by JobsCreationSucceeded

andreyvelich · 2024-11-08T15:27:19Z

pkg/controller.v2/trainjob_controller.go

+	succeeded      objsOpState = iota
+	buildFailed    objsOpState = iota
+	creationFailed objsOpState = iota
+	updateFailed   objsOpState = iota


What is the goal to represent conditions as increment int constants ? Does it require less memory than just using strings ?

This is Go World ENUM. Shouldn't the state be represented as Enums?
Additionally, the string could be unhandled since the string has loose constraints.
It would be better to manage states as strong constraints.

I see, do you know if k8s uses the same representation for the conditions ?

do you know if k8s uses the same representation for the conditions ?

Does this mean that whether or not any K8s controllers (like Deployment Controller) have the same transition calculation mechanism?

andreyvelich · 2024-11-08T15:32:44Z

pkg/controller.v2/trainjob_controller.go

+	if !equality.Semantic.DeepEqual(&trainJob, originStatus) {
+		return ctrl.Result{}, errors.Join(err, r.client.Status().Update(ctx, &trainJob))


Why do we return error if objects are not equal and TrainJob condition needs to be updated ?

This error will be returned only when err or r.client.Status().Update(ctx, &trainJob) has errors.
So, when both do not have any error, this will return nil.

Oh, you are right!

andreyvelich · 2024-11-08T15:34:09Z

pkg/controller.v2/trainjob_controller.go

+type objsOpState int
+
+const (
+	succeeded      objsOpState = iota


Let's call it creationSucceeded to be more explicit ?

Suggested change

succeeded objsOpState = iota

creationSucceeded objsOpState = iota

andreyvelich · 2024-11-08T15:40:36Z

pkg/runtime.v2/framework/plugins/jobset/jobset.go

+		return nil, err
+	}
+	if completed := meta.FindStatusCondition(jobSet.Status.Conditions, string(jobsetv1alpha2.JobSetCompleted)); completed != nil && completed.Status == metav1.ConditionTrue {
+		completed.Reason = fmt.Sprintf("%sDueTo%s", completed.Type, completed.Reason)


The Reason can't include spaces ?
E.g. here we say that FailedDueTo...
Should we use Message for more detailed explantation ?
For example:

Type: Failed Reason: JobSetReachedMaxRestarts Message: jobset failed due to reaching max number of restarts <- Message from JobSet

The reason is machine-readable field, and should be CamelCase.

Additionally, As you can see in the following, we already set the JobSet message:

training-operator/test/integration/controller.v2/trainjob_controller_test.go

Line 426 in d5230f7

Message: jobsetconsts.AllJobsCompletedMessage,

That makes sense. Why you don't want to just transition reason from the failed/completed JobSet to the TrainJob reason ?

I was wondering if other arbitrary Job CRDs have less context Reason like just "Finished".
So, I connected Type and Reason by the DueTo so that we can use the same form Reason in all Job CRDs.
However, the TerminalCondition will be implemented by the external developer, and then they can return arbitrary conditions.

As reconsidering this, we can just directly use the JobSet Reason without connecting with Type since we do not have any guarantees at an interface level that the plugin provides a uniform format.

I revised this one.

pkg/runtime.v2/framework/plugins/jobset/jobset.go

andreyvelich · 2024-11-08T15:49:38Z

pkg/controller.v2/trainjob_controller.go

+		newCond = metav1.Condition{
+			Type:    kubeflowv2.TrainJobCreated,
+			Status:  metav1.ConditionTrue,
+			Message: "Succeeded to create Jobs",


We might need to move messages to the Constants as well.

Signed-off-by: Yuki Iwai <[email protected]>

tenzen-y · 2024-11-09T00:04:07Z

@andreyvelich I addressed all comments. PTAL, thanks!

Signed-off-by: Yuki Iwai <[email protected]>

andreyvelich · 2024-11-09T14:35:45Z

Thanks @tenzen-y!
/lgtm
/approve
/hold

Feel free to merge it.

google-oss-prow · 2024-11-09T14:35:52Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: andreyvelich

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

~~OWNERS~~ [andreyvelich]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

tenzen-y · 2024-11-09T18:13:38Z

Thank you for the review!
/hold cancel

* KEP-2170: Implement TrainJob conditions Signed-off-by: Yuki Iwai <[email protected]> * Fix API comments Signed-off-by: Yuki Iwai <[email protected]> * Make condition message constants Signed-off-by: Yuki Iwai <[email protected]> * Stop connecting condition type and reason in JobSet plugin Signed-off-by: Yuki Iwai <[email protected]> --------- Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]>

* Added test for create-pytorchjob.ipynb Signed-off-by: sailesh duddupudi <[email protected]> * fix yaml syntax Signed-off-by: sailesh duddupudi <[email protected]> * Fix uses path Signed-off-by: sailesh duddupudi <[email protected]> * Add actions/checkout Signed-off-by: sailesh duddupudi <[email protected]> * Add bash to action.yaml Signed-off-by: sailesh duddupudi <[email protected]> * Install pip dependencies step Signed-off-by: sailesh duddupudi <[email protected]> * Add quotes for args Signed-off-by: sailesh duddupudi <[email protected]> * Add jupyter Signed-off-by: sailesh duddupudi <[email protected]> * Add nbformat_minor: 5 to fix invalid format error Signed-off-by: sailesh duddupudi <[email protected]> * Fix job name Signed-off-by: sailesh duddupudi <[email protected]> * test papermill-args-yaml Signed-off-by: sailesh duddupudi <[email protected]> * testing multi line args Signed-off-by: sailesh duddupudi <[email protected]> * testing multi line args1 Signed-off-by: sailesh duddupudi <[email protected]> * testing multi line args2 Signed-off-by: sailesh duddupudi <[email protected]> * testing multi line args3 Signed-off-by: sailesh duddupudi <[email protected]> * Parameterize sdk install Signed-off-by: sailesh duddupudi <[email protected]> * Remove unnecessary output Signed-off-by: sailesh duddupudi <[email protected]> * nbformat normailze Signed-off-by: sailesh duddupudi <[email protected]> * [SDK] Training Client Conditions related unit tests (#2253) * test: add unit test for get_job_conditions function of training client Signed-off-by: Bobbins228 <[email protected]> * test: add unit test for is_job_created function of training client Signed-off-by: Bobbins228 <[email protected]> * test: add unit test for is_job_running function of training client Signed-off-by: Bobbins228 <[email protected]> * test: add unit test for is_job_restarting function of training client Signed-off-by: Bobbins228 <[email protected]> * test: add unit test for is_job_failed function of training client Signed-off-by: Bobbins228 <[email protected]> * test: add unit test for is_job_succeded function of training client Signed-off-by: Bobbins228 <[email protected]> * test: improve job condition unit tests efficiency Signed-off-by: Bobbins228 <[email protected]> --------- Signed-off-by: Bobbins228 <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * [SDK] test: add unit test for list_jobs method of the training_client (#2267) Signed-off-by: wei-chenglai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Generate clientset, openapi spec for the V2 APIs (#2273) Generate clientset, informers, listers and open api spec for v2alpha1 APIs. Signed-off-by: Varsha Prasad Narsing <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * [SDK] Use torchrun to create PyTorchJob from function (#2276) * [SDK] Use torchrun to create PyTorchJob from function Signed-off-by: Andrey Velichkevich <[email protected]> * Update PyTorchJob SDK example Signed-off-by: Andrey Velichkevich <[email protected]> * Add consts for entrypoint Signed-off-by: Andrey Velichkevich <[email protected]> * Add check for num procs per worker Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * [SDK] test: add unit test for get_job_logs method of the training_client (#2275) Signed-off-by: wei-chenglai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * [v2alpha] Move GV related codebase (#2281) Move GV related codebase in v2alpha Signed-off-by: Varsha Prasad Narsing <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Implement runtime framework (#2248) * KEP-2170: Implement runtime framework interfaces Signed-off-by: Yuki Iwai <[email protected]> * Remove grep dependency Signed-off-by: Yuki Iwai <[email protected]> * KEP-2170: Implement ValidateObjects interface to the runtime framework Signed-off-by: Yuki Iwai <[email protected]> * KEP-2170: Expose the TrainingRuntime and ClusterTrainingRuntime Kind Signed-off-by: Yuki Iwai <[email protected]> * KEP-2170: Remove unneeded scheme field from the internal TrainingRuntime Signed-off-by: Yuki Iwai <[email protected]> * Rephrase the error message Signed-off-by: Yuki Iwai <[email protected]> * Distinguish TrainingRuntime and ClusterTrainingRuntime when creating indexes for the TrainJobs Signed-off-by: Yuki Iwai <[email protected]> * Propagate the TrainJob labels and annotations to the JobSet Signed-off-by: Yuki Iwai <[email protected]> * Remove PodAnnotations from the runtime info Signed-off-by: Yuki Iwai <[email protected]> * Implement TrainingRuntime ReplicatedJob validation Signed-off-by: Yuki Iwai <[email protected]> * Add TODO comments Signed-off-by: Yuki Iwai <[email protected]> * Replace queueSuspendedTrainJob with queueSuspendedTrainJobs Signed-off-by: Yuki Iwai <[email protected]> --------- Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Add DeepSpeed Example with Pytorch Operator (#2235) Signed-off-by: Syulin7 <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Rename TrainingRuntimeRef to RuntimeRef API (#2283) * KEP-2170: Rename TrainingRuntimeRef to RuntimeRef API Signed-off-by: Andrey Velichkevich <[email protected]> * Rename RuntimeRef in runtime framework Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Adding CEL validations on v2 TrainJob CRD (#2260) Signed-off-by: Akshay Chitneni <[email protected]> Co-authored-by: Akshay Chitneni <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Upgrade Deepspeed demo dependencies (#2294) Signed-off-by: Syulin7 <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Add manifests for Kubeflow Training V2 (#2289) * KEP-2170: Add manifests for Kubeflow Training V2 Signed-off-by: Andrey Velichkevich <[email protected]> * Fix invalid name for webhook config in cert Signed-off-by: Andrey Velichkevich <[email protected]> * Fix integration tests Signed-off-by: Andrey Velichkevich <[email protected]> * Move kubebuilder markers to runtime framework Signed-off-by: Andrey Velichkevich <[email protected]> * Use Kubernetes recommended labels Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * FSDP Example for T5 Fine-Tuning and PyTorchJob (#2286) * FSDP Example with PyTorchJob and T5 Fine-Tuning Signed-off-by: Andrey Velichkevich <[email protected]> * Modify text Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Implement TrainJob Reconciler to manage objects (#2295) * KEP-2170: Implement TrainJob Reconciler to manage objects Signed-off-by: Yuki Iwai <[email protected]> * Mode dep-crds to manifests/external-crds Signed-off-by: Yuki Iwai <[email protected]> * Rename run with runtime Signed-off-by: Yuki Iwai <[email protected]> --------- Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Remove Prometheus Monitoring doc (#2301) Signed-off-by: Sophie <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Decouple JobSet from TrainJob (#2296) Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Strictly verify the CRD marker validation and defaulting in the integration testings (#2304) Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Initialize runtimes before the manager starts (#2306) Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Generate Python SDK for Kubeflow Training V2 (#2310) * Generate SDK models for the Training V2 APIs Signed-off-by: Andrey Velichkevich <[email protected]> * Create pyproject.toml config Signed-off-by: Andrey Velichkevich <[email protected]> * Remove comments Signed-off-by: Andrey Velichkevich <[email protected]> * Fix pre-commit Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Create model and dataset initializers (#2303) * KEP-2170: Create model and dataset initializers Signed-off-by: Andrey Velichkevich <[email protected]> * Add abstract classes Signed-off-by: Andrey Velichkevich <[email protected]> * Add storage URI to config Signed-off-by: Andrey Velichkevich <[email protected]> * Update .gitignore Co-authored-by: Kevin Hannon <[email protected]> Signed-off-by: Andrey Velichkevich <[email protected]> * Fix the misspelling for initializer Signed-off-by: Andrey Velichkevich <[email protected]> * Add .pt and .pth to ignore_patterns Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Co-authored-by: Kevin Hannon <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Implement JobSet, PlainML, and Torch Plugins (#2308) * KEP-2170: Implement JobSet and PlainML Plugins Signed-off-by: Andrey Velichkevich <[email protected]> * Fix nil pointer exception for Trainer Signed-off-by: Andrey Velichkevich <[email protected]> * Fix unit tests in runtime package Signed-off-by: Andrey Velichkevich <[email protected]> * Fix unit tests Signed-off-by: Andrey Velichkevich <[email protected]> * Fix integration tests Signed-off-by: Andrey Velichkevich <[email protected]> * Fix lint Signed-off-by: Andrey Velichkevich <[email protected]> * Implement Torch Plugin Signed-off-by: Andrey Velichkevich <[email protected]> * Use list for the Info envs Signed-off-by: Andrey Velichkevich <[email protected]> * Fix golang ci Signed-off-by: Andrey Velichkevich <[email protected]> * Fix Torch plugin Signed-off-by: Andrey Velichkevich <[email protected]> * Use K8s sets Update error return Use ptr.Deref() for nil values Signed-off-by: Andrey Velichkevich <[email protected]> * Use client.Object for Build() call Signed-off-by: Andrey Velichkevich <[email protected]> * Remove DeepCopy Signed-off-by: Andrey Velichkevich <[email protected]> * Remove MLPolicy and PodGroupPolicy from the Info object Signed-off-by: Andrey Velichkevich <[email protected]> * Inline error Signed-off-by: Andrey Velichkevich <[email protected]> * Remove SDK jar file Signed-off-by: Andrey Velichkevich <[email protected]> * Add integration test for Torch plugin Signed-off-by: Andrey Velichkevich <[email protected]> * Add TODO to calculate PodGroup values in unit tests Signed-off-by: Andrey Velichkevich <[email protected]> * Revert the change to add original Runtime Policies to Info Signed-off-by: Andrey Velichkevich <[email protected]> * Create const for the DefaultJobReplicas Signed-off-by: Andrey Velichkevich <[email protected]> * Check if PodLabels is empty Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Implement Initializer builders in the JobSet plugin (#2316) * KEP-2170: Implement Initializer builder in the JobSet plugin Signed-off-by: Andrey Velichkevich <[email protected]> * Update the SDK models Signed-off-by: Andrey Velichkevich <[email protected]> * Remove Info from Initializer builder Signed-off-by: Andrey Velichkevich <[email protected]> * Update manifests Signed-off-by: Andrey Velichkevich <[email protected]> * Update pkg/constants/constants.go Co-authored-by: Yuki Iwai <[email protected]> Signed-off-by: Andrey Velichkevich <[email protected]> * Use var for envs Signed-off-by: Andrey Velichkevich <[email protected]> * Remove check manifests from GitHub actions Signed-off-by: Andrey Velichkevich <[email protected]> * Move consts to JobSet plugin Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Co-authored-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Add the TrainJob state transition design (#2298) * KEP-2170: Add the TrainJob state transition design Signed-off-by: Yuki Iwai <[email protected]> * Replace actual jobs with TrainJob Signed-off-by: Yuki Iwai <[email protected]> * Remove the JobSet conditions propagation and Add expanding runtime framework interfaces for each plugin Signed-off-by: Yuki Iwai <[email protected]> * Expand the Creation Failed reasons Signed-off-by: Yuki Iwai <[email protected]> * Rename Completed to Complete Signed-off-by: Yuki Iwai <[email protected]> --------- Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Update tf job examples to tf v2 (#2270) * mnist with summaries updaetd to TF v2 Signed-off-by: yelias <[email protected]> * tf_sample updaetd to TF v2 Signed-off-by: yelias <[email protected]> * Add mnist_utils and update dist-mnist Signed-off-by: yelias <[email protected]> * Add mnist_utils and update dist-mnist Signed-off-by: yelias <[email protected]> * Remove old example - estimator-API, this example has been replaced by distribution_strategy Signed-off-by: yelias <[email protected]> * Small fix Signed-off-by: yelias <[email protected]> * Remove unsupported powerPC dockerfiles Signed-off-by: yelias <[email protected]> * Fix typo in copyright Signed-off-by: yelias <[email protected]> --------- Signed-off-by: yelias <[email protected]> Co-authored-by: yelias <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Add TrainJob conditions (#2322) * KEP-2170: Implement TrainJob conditions Signed-off-by: Yuki Iwai <[email protected]> * Fix API comments Signed-off-by: Yuki Iwai <[email protected]> * Make condition message constants Signed-off-by: Yuki Iwai <[email protected]> * Stop connecting condition type and reason in JobSet plugin Signed-off-by: Yuki Iwai <[email protected]> --------- Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Pin Gloo repository in JAX Dockerfile to a specific commit (#2329) This commit pins the Gloo repository to a specific commit (43b7acbf) in the JAX Dockerfile to prevent build failures caused by a recent bug introduced in the Gloo codebase. By locking the version of Gloo to a known working commit, we ensure that the JAX build remains stable and functional until the issue is resolved upstream. The build failure occurs when compiling the gloo/transport/tcp/buffer.cc file due to an undefined __NR_gettid constant, which was introduced after the pinned commit. By using this commit, we bypass the issue and allow the build to complete successfully. Signed-off-by: Sandipan Panda <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * [fix] Resolve v2alpha API exceptions (#2317) Resolve v2alpha API exceptions by adding necessary listType validations. Signed-off-by: Varsha Prasad Narsing <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Upgrade Kubernetes to v1.30.7 (#2332) * Upgrade Kubernetes to v1.30.7 Signed-off-by: Antonin Stefanutti <[email protected]> * Use typed event handlers and predicates in job controllers Signed-off-by: Antonin Stefanutti <[email protected]> * Re-organize pkg/common/util/reconciler.go Signed-off-by: Antonin Stefanutti <[email protected]> * Update installation instructions in README Signed-off-by: Antonin Stefanutti <[email protected]> --------- Signed-off-by: Antonin Stefanutti <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Ignore cache exporting errors in the image building workflows (#2336) Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * KEP-2170: Add Torch Distributed Runtime (#2328) * KEP-2170: Add Torch Distributed Runtime Signed-off-by: Andrey Velichkevich <[email protected]> * Add pip list Signed-off-by: Andrey Velichkevich <[email protected]> --------- Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Refine the server-side apply installation args (#2337) Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Add openapi-generator CLI option to skip SDK v2 test generation (#2338) Signed-off-by: Antonin Stefanutti <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Upgrade kustomization files to Kustomize v5 (#2326) Signed-off-by: oksanabaza <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Pin accelerate package version in trainer (#2340) * Pin accelerate package version in trainer Signed-off-by: Gavrish Prabhu <[email protected]> * include new line to pass pre-commit hook Signed-off-by: Gavrish Prabhu <[email protected]> --------- Signed-off-by: Gavrish Prabhu <[email protected]> Signed-off-by: sailesh duddupudi <[email protected]> * Replace papermill command with bash script Signed-off-by: sailesh duddupudi <[email protected]> * Typo fix Signed-off-by: sailesh duddupudi <[email protected]> * Move Checkout step outside action.yaml file Signed-off-by: sailesh duddupudi <[email protected]> * Add newline EOF in script Signed-off-by: sailesh duddupudi <[email protected]> * Pass python dependencies as args and pin versions Signed-off-by: sailesh duddupudi <[email protected]> * Update Usage Signed-off-by: sailesh duddupudi <[email protected]> * Install dependencies in yaml Signed-off-by: sailesh duddupudi <[email protected]> * fix ipynb Signed-off-by: sailesh duddupudi <[email protected]> * set bash flags Signed-off-by: sailesh duddupudi <[email protected]> * Update script args and add more kubernetes versions for tests Signed-off-by: sailesh duddupudi <[email protected]> * add gang-scheduler-name to template Signed-off-by: sailesh duddupudi <[email protected]> * move go setup to template Signed-off-by: sailesh duddupudi <[email protected]> * remove -p parameter from script Signed-off-by: sailesh duddupudi <[email protected]> --------- Signed-off-by: sailesh duddupudi <[email protected]> Signed-off-by: Bobbins228 <[email protected]> Signed-off-by: wei-chenglai <[email protected]> Signed-off-by: Varsha Prasad Narsing <[email protected]> Signed-off-by: Andrey Velichkevich <[email protected]> Signed-off-by: Yuki Iwai <[email protected]> Signed-off-by: Syulin7 <[email protected]> Signed-off-by: Akshay Chitneni <[email protected]> Signed-off-by: Sophie <[email protected]> Signed-off-by: yelias <[email protected]> Signed-off-by: Sandipan Panda <[email protected]> Signed-off-by: Antonin Stefanutti <[email protected]> Signed-off-by: oksanabaza <[email protected]> Signed-off-by: Gavrish Prabhu <[email protected]> Co-authored-by: Mark Campbell <[email protected]> Co-authored-by: Wei-Cheng Lai <[email protected]> Co-authored-by: Varsha <[email protected]> Co-authored-by: Andrey Velichkevich <[email protected]> Co-authored-by: Yuki Iwai <[email protected]> Co-authored-by: yu lin <[email protected]> Co-authored-by: Akshay Chitneni <[email protected]> Co-authored-by: Akshay Chitneni <[email protected]> Co-authored-by: Sophie Hsu <[email protected]> Co-authored-by: Kevin Hannon <[email protected]> Co-authored-by: YosiElias <[email protected]> Co-authored-by: yelias <[email protected]> Co-authored-by: Sandipan Panda <[email protected]> Co-authored-by: Antonin Stefanutti <[email protected]> Co-authored-by: Oksana Bazylieva <[email protected]> Co-authored-by: Gavrish Prabhu <[email protected]>

google-oss-prow bot added the do-not-merge/work-in-progress label Nov 7, 2024

google-oss-prow bot requested review from jinchihe and kuizhiqing November 7, 2024 14:47

google-oss-prow bot added the size/L label Nov 7, 2024

tenzen-y force-pushed the v2-add-reconcile-condition branch from 128a039 to e1bd573 Compare November 7, 2024 20:43

google-oss-prow bot added size/XL and removed size/L labels Nov 7, 2024

tenzen-y force-pushed the v2-add-reconcile-condition branch 2 times, most recently from 8c638b4 to fd61ff2 Compare November 7, 2024 20:45

tenzen-y marked this pull request as ready for review November 7, 2024 20:46

tenzen-y force-pushed the v2-add-reconcile-condition branch from fd61ff2 to 8ca03df Compare November 7, 2024 20:48

tenzen-y commented Nov 7, 2024

View reviewed changes

tenzen-y force-pushed the v2-add-reconcile-condition branch from 8ca03df to 665b605 Compare November 7, 2024 20:49

tenzen-y force-pushed the v2-add-reconcile-condition branch 2 times, most recently from 4bb08b7 to 12514ab Compare November 7, 2024 20:52

tenzen-y changed the title ~~WIP: KEP-2170: Add TrainJob conditions~~ KEP-2170: Add TrainJob conditions Nov 7, 2024

google-oss-prow bot removed the do-not-merge/work-in-progress label Nov 7, 2024

google-oss-prow bot added the do-not-merge/hold label Nov 7, 2024

KEP-2170: Implement TrainJob conditions

d5230f7

Signed-off-by: Yuki Iwai <[email protected]>

tenzen-y force-pushed the v2-add-reconcile-condition branch from 12514ab to d5230f7 Compare November 7, 2024 22:48

andreyvelich reviewed Nov 8, 2024

View reviewed changes

tenzen-y added 2 commits November 9, 2024 08:34

Fix API comments

79ff67b

Signed-off-by: Yuki Iwai <[email protected]>

Make condition message constants

58b1a00

Signed-off-by: Yuki Iwai <[email protected]>

Stop connecting condition type and reason in JobSet plugin

c36fe90

Signed-off-by: Yuki Iwai <[email protected]>

tenzen-y force-pushed the v2-add-reconcile-condition branch from b2cc1ef to c36fe90 Compare November 9, 2024 07:00

google-oss-prow bot assigned andreyvelich Nov 9, 2024

google-oss-prow bot added the lgtm label Nov 9, 2024

google-oss-prow bot added the approved label Nov 9, 2024

google-oss-prow bot removed the do-not-merge/hold label Nov 9, 2024

google-oss-prow bot merged commit 94dee0e into kubeflow:master Nov 9, 2024
42 checks passed

tenzen-y deleted the v2-add-reconcile-condition branch November 9, 2024 18:14

	// TrainJobSuspended means the TrainJob is suspended.
	// TrainJobSuspended means that the TrainJob is suspended.

		if !equality.Semantic.DeepEqual(&trainJob, originStatus) {
		return ctrl.Result{}, errors.Join(err, r.client.Status().Update(ctx, &trainJob))

	succeeded objsOpState = iota
	creationSucceeded objsOpState = iota

KEP-2170: Add TrainJob conditions #2322

KEP-2170: Add TrainJob conditions #2322

Conversation

tenzen-y commented Nov 7, 2024 • edited Loading

Choose a reason for hiding this comment

coveralls commented Nov 7, 2024 • edited Loading

Pull Request Test Coverage Report for Build 11754225694

Details

💛 - Coveralls

tenzen-y commented Nov 7, 2024

tenzen-y commented Nov 7, 2024

andreyvelich left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

andreyvelich Nov 9, 2024 • edited Loading

Choose a reason for hiding this comment

andreyvelich Nov 8, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tenzen-y Nov 8, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

andreyvelich Nov 9, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tenzen-y commented Nov 9, 2024

andreyvelich commented Nov 9, 2024

google-oss-prow bot commented Nov 9, 2024

tenzen-y commented Nov 9, 2024

tenzen-y commented Nov 7, 2024 •

edited

Loading

coveralls commented Nov 7, 2024 •

edited

Loading

andreyvelich left a comment •

edited

Loading

andreyvelich Nov 9, 2024 •

edited

Loading

andreyvelich Nov 8, 2024 •

edited

Loading

tenzen-y Nov 8, 2024 •

edited

Loading

andreyvelich Nov 9, 2024 •

edited

Loading