feat(validator): add Kubeflow Trainer to robust-controller and skip inference-gateway on training clusters#349
Merged
yuanchen8911 merged 1 commit intoNVIDIA:mainfrom Mar 12, 2026
Conversation
1ae1119 to
60c6076
Compare
60c6076 to
0972e98
Compare
dims
previously approved these changes
Mar 11, 2026
0972e98 to
1f0f531
Compare
1f0f531 to
73f5395
Compare
73f5395 to
2b9392a
Compare
51d0372 to
7e21445
Compare
7e21445 to
3e0586b
Compare
dims
approved these changes
Mar 11, 2026
3e0586b to
d2b47f9
Compare
The robust-controller conformance check previously only validated the Dynamo operator, causing it to skip on all training clusters. This adds Kubeflow Trainer as an alternative target, selected based on recipe component presence: - dynamo-platform in recipe → validate Dynamo operator - kubeflow-trainer in recipe → validate Kubeflow Trainer - neither → skip Kubeflow Trainer validation checks: 1. Controller deployment running (kubeflow-trainer-controller-manager) 2. Validating webhook operational with reachable endpoint 3. TrainJob CRD exists (trainjobs.trainer.kubeflow.org) 4. Webhook rejects invalid TrainJob (behavioral test) Refactored the original Dynamo validation into checkRobustDynamo() and renamed validateWebhookRejects to validateDynamoWebhookRejects for clarity.
d2b47f9 to
3a0c795
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
robust-controllerCNCF conformance checkinference-gatewaycheck on training clusters (no kgateway)collect-evidence.sh)Problem
Changes
robust-controller (Go + evidence script)
Check selects target based on recipe component presence:
dynamo-platformin recipe → validate Dynamo operator (existing)kubeflow-trainerin recipe → validate Kubeflow Trainer (new)Kubeflow Trainer validation:
kubeflow-trainer-controller-manager)validator.trainer.kubeflow.orgtrainjobs.trainer.kubeflow.org)Evidence script (
collect-evidence.sh):inference-gateway (Go + evidence script)
kgatewaycomponent is not in recipe (training clusters)Test plan
TestRecipeHasComponent(5 cases),TestCheckRobustControllerRouting(6 cases)go build,go vet,gofmtclean