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
We need a way to monitor the continuous delivery of Kubeflow applications.
As a strawman I think we'd like a dashboard that shows a table like the following for each Kubeflow version.
Application
Last Commit of Code
Image Commit
Status of Update Pipeline
Notebook Controller
1234
1234
Green
Profile Controller
abcd
efgh
Red
...
...
...
...
So by looking at the dashboard we'd like to be able to answer the following questions
For Kubeflow version x.y.z which commit of application A is in the manifest?
Is the manifest for Application A up to date?
Are the CD pipelines for Application A failing?
We could probably easily write a python script to compute the above information. We could then perhaps turn that into a report just by having a flask server return it as html.
Perhaps we should start by writing a notebook and then think about how to turn that notebook into
reports being served at some URL?
The text was updated successfully, but these errors were encountered:
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in one week if no further activity occurs. Thank you for your contributions.
Follow on to #450
We need a way to monitor the continuous delivery of Kubeflow applications.
As a strawman I think we'd like a dashboard that shows a table like the following for each Kubeflow version.
So by looking at the dashboard we'd like to be able to answer the following questions
We could probably easily write a python script to compute the above information. We could then perhaps turn that into a report just by having a flask server return it as html.
Perhaps we should start by writing a notebook and then think about how to turn that notebook into
reports being served at some URL?
The text was updated successfully, but these errors were encountered: