Closed streamnsight closed 5 months ago
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
This issue has been automatically closed because it has not had recent activity. Please comment "/reopen" to reopen it.
There has been PRs to fix the delete function in launch_tfjob but the image has not been updated in the component
This PR https://github.com/kubeflow/pipelines/issues/7984 fixes the launch_crd.py, but the component here: https://github.com/kubeflow/pipelines/blob/master/components/kubeflow/launcher/component.yaml references the image
nikenano/launchernew:latest
This image is over 2 years old and still has the bug: https://hub.docker.com/r/nikenano/launchernew/tags
From the repo, the build_image.sh script reference the local image as
ml-pipeline-kubeflow-tfjob
, and the corresponding imagegcr.io/ml-pipeline/ml-pipeline-kubeflow-tfjob:1.8.5
does not work either; it does not have the code packaged installed under/ml
properly, it has/ml/src/launch_tfjob.py
and/ml/common/launch_crd.py
When running it I get:
running the launch_tfjob.py from the /ml/src/ dir does not work either because the launch_crd is not in the right folder structure.
The image needs to be rebuilt and pushed, and the component.yaml needs to be updated with the latest image.
Impacted by this bug? Give it a 👍.