Closed tenzen-y closed 1 year ago
@johnugeorge Can we wait to create a new kubeflow/common release until this PR is merged?
This means when hpa-controller updates
replcase
, training-operator does not update PodGroup.
are you referring replicas
?
This means when hpa-controller updates
replcase
, training-operator does not update PodGroup.are you referring
replicas
?
Ah, yes. This is a typo.
/lgtm
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: terrytangyuan
The full list of commands accepted by this bot can be found here.
The pull request process is described here
Signed-off-by: Yuki Iwai yuki.iwai.tz@gmail.com
Currently, when users update CustomJob resources (e.g, TFJob) with
runPolicy.schedulingPolicy
orreplicas
changes, traininig-operator can not update PodGroup.This means when hpa-controller updates
replicas
, training-operator does not update PodGroup.So I implemented the logic to update PodGroup.
/assign @zw0610 @johnugeorge @terrytangyuan