You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm using Kubeflow with on-premise Kubernetes cluster.
As we use Rancher local path provisioner for Storage Class it is crucial to control disk usage (this class does not work with pvc limits).
I wonder if it is possible to set some kind of period of validity for each artifact to periodically and automatically delete the oldest.
Is there a workaround currently?
I was thinking of preparing k8s crone job to do it manually, but I don't know specifically pipelines architecture and already noticed that deletion from minio is not enough (cacher still finds deleted artifacts available).
Love this idea? Give it a 👍. We prioritize fulfilling features with the most 👍.
The text was updated successfully, but these errors were encountered:
Hello @zijianjoy, thank you for your response, I am aware about minio rules, but I think it will not work very well as cacher will still find those expired artifacts as available.
Feature Area
/area backend
/area components
What feature would you like to see?
Pipeline run artifacts validity period.
What is the use case or pain point?
I'm using Kubeflow with on-premise Kubernetes cluster.
As we use Rancher local path provisioner for Storage Class it is crucial to control disk usage (this class does not work with pvc limits).
I wonder if it is possible to set some kind of period of validity for each artifact to periodically and automatically delete the oldest.
Is there a workaround currently?
I was thinking of preparing k8s crone job to do it manually, but I don't know specifically pipelines architecture and already noticed that deletion from minio is not enough (cacher still finds deleted artifacts available).
Love this idea? Give it a 👍. We prioritize fulfilling features with the most 👍.
The text was updated successfully, but these errors were encountered: