sanchesoon

@thesuperzapper It give us ability to use Virtual Host style. Not sure if here a lot work, should be just additional parameter to pass launcher (but I never did PR...

Hi @HumairAK What do you think about this ? Points: - DS engineer would be interesting to see logs as artifact, as part of models producer investigation - logs can...

Why engineers may interesting in this: - some frameworks for ML logging kind of metrics (I took this picture from the Internet )  - kubeflow run is snapshot of...

> > Current workaround with argoworkflow backend is using keyfomat like /artifacts/{workflow_id}/{pod_id}/main.log via patching artifact_ui split function and use v1 handler. > > @sanchesoon Can you give more details on...

Hi all, any plans to merge it ?

For argoworkflow backend I am using something like this ``` mainContainer: | imagePullPolicy: IfNotPresent resources: requests: cpu: 0.1 memory: 64Mi limits: cpu: 0.5 memory: 512Mi ports: - containerPort: 8080 ```...

As workaround you can do python wrapper for component. Real problem caused in api-server + argoworkflow. There are mismatch between names.