-
Bug
-
Resolution: Done
-
Major
-
None
-
None
-
2
-
False
-
-
False
-
None
-
Testable
-
No
-
-
-
-
-
-
-
No
-
No
-
Pending
-
None
-
-
Description of problem:
In an older DS project, many of my workbenches now have an "unknown" image associated:
If I try to edit the project, I can't see which image it was using, I'm forced to choose from the available images.
And yet, I can still start and use this workbench.
Prerequisites (if any, like setup, operators/versions):
Have an older cluster with projects that were created a while back.
Steps to Reproduce
- <steps>
Actual results:
- unknown image.
- unable to know which one it even was. (was it a tensorflow or a pytorch?)
Expected results:
- at least know which image it was.
- even if I can't select an N-2 image for a fresh project, I should be able to keep using it.
Reproducibility (Always/Intermittent/Only Once):
I've seen the same thing in other projects too.
Build Details:
Workaround:
Additional info:
These are the images behind the scenes:
bash-4.4 ~ $ oc get statefulsets -n a-great-project -o jsonpath='{range .items[*]}{.spec.template.spec.containers[*].image}{"\n"}{end}' image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/byon-1681497173054:jupyter-datascience-ubi9-python-3.9-20230411 image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/pytorch:py3.8-cuda-11.4.2-2 image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/byon-1679684285597:latest image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/tensorflow:py3.8-cuda-11.4.2-2 image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/s2i-minimal-notebook:py3.8-v1 image-registry.openshift-image-registry.svc:5000/redhat-ods-applications/byon-1675352918609:code-server-ubi9-py39_2023b_latest bash-4.4 ~ $