search

kubeflow / kfp-tekton

Kubeflow Pipelines on Tekton
https://developer.ibm.com/blogs/kubeflow-pipelines-with-tekton-and-watson/
Apache License 2.0
175 stars 122 forks source link

Server API does not accept Tekton YAML with dependency specified by `after` on condition node #310

Closed rafalbigaj closed 4 years ago

rafalbigaj commented 4 years ago

/kind bug

What steps did you take and what happened:

The successfully compiled tekton YAML: train_and_validate_in_notebook_pipeline.py.tar.gz

from source: https://github.ibm.com/AILifecycle/ai-kubeflow-pipelines/blob/new-components/pipelines/train_and_validate_in_notebook_pipeline.py#L135

fails, when trying to create a run with error:

 InternalServerError: Failed to create a workflow for (train-the-model-and-evaluate-in-notebook-f3ee6):
 admission webhook \"validation.webhook.pipeline.tekton.dev\" denied the request: validation failed: invalid value:
 couldn't add link between run-notebook and condition-positive-condition-1:
 task run-notebook depends on condition-positive-condition-1 but condition-positive-condition-1 wasn't present in Pipeline: 
 spec.tasks

What did you expect to happen:

The same DSL code compiled to Argo YAML can be run without any issues.

Additional information: N/A

Environment:

issue-label-bot[bot] commented 4 years ago

Issue Label Bot is not confident enough to auto-label this issue. See dashboard for more details.

issue-label-bot[bot] commented 4 years ago

Issue Label Bot is not confident enough to auto-label this issue. See dashboard for more details.

Tomcli commented 4 years ago

Thanks @rafalbigaj, we will take a look and see can we do some workaround with the alpha conditional API. If not, we will prompt an error in the SDK for now and try to address it in the beta conditional API during the next major release.

Tomcli commented 4 years ago

FYI @ckadner @drewbutlerbb4 Since the alpha tekton condition is not a regular task (also it's an older api), it's not possible to refer it as a dependencies such as .after(conditions) or condition.results. So let's see is there any workaround we can do on the compiler side.

Tomcli commented 4 years ago

This feature is supported in the master branch now.

Tomcli commented 4 years ago

/close

k8s-ci-robot commented 4 years ago

@Tomcli: Closing this issue.

In response to [this](https://github.com/kubeflow/kfp-tekton/issues/310#issuecomment-730049546): >/close Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.