WIP: support worker and metricsCollector template spec #375
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: If they are not already assigned, you can assign the PR to them by writing The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
hougangliu
changed the title
| wtp, err = template.New("Worker").Parse(wt) | ||
| tName, tNamespace, tPath := getDefaultTemplateSpec() | ||
| if goTemplate.TemplateSpec != nil { | ||
| tName = goTemplate.TemplateSpec.ConfigMapName |
There was a problem hiding this comment.
When the field of TemplateSpec is emply, we should use the default value.
For example, we can omit the namespace of ConfigMap when we want to use the same namespace as the studyjobcontroller.
| @@ -114,19 +114,25 @@ const ( | |||
| OptimizationTypeMaximize OptimizationType = "maximize" | |||
| ) | |||
|
|
|||
| type GoTemplate struct { | |||
| type TemplateSpec struct { | |||
There was a problem hiding this comment.
This extension of studyjob API would be for v1alpha2? #370
|
This issue is listed v1alpha2 API #370 . |
|
@hougangliu: The following test failed, say
Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
|
Yes. I would suggest that only common changes(that won't break the current config) go into v1alpha1 at this point. @richardsliu is working to get v1alpha2 structure. Once it is merged, we can get it in v1alpha2 In general, I feel that we are moving away from the data scientist point of view, instead thinking from a kubernetes developer perspective. Currently, a data scientist will find it extremely difficult to write a katib study job config correctly without good kubernetes experience. For eg: Currently in the studyjob config, there is a namespace for overall studyjob, a namespace for the workerspec, namespace for ConfigMap, namespace for metric collector . Won't this be complicated for the users by exposing much details? WDYT? |
|
will submit another PR for v1alpha2 |
Fixes: #349
This change is