unschedulablePodWithGpuTimeBuffer configuration

[maxpain]

Hello. How do you configure this variable?

autoscaler/cluster-autoscaler/core/static_autoscaler.go

Line 69 in 57f2814

unschedulablePodWithGpuTimeBuffer = 30 * time.Second

NVIDIA L4 spot GPUs in GCE are cheap for our use case, so we need to scale up immediately.

[vadasambar]

@maxpain I don't think the variable is configurable right now. We might need to expose it via flag if you want to configure it.

[aiden-von]

May I handle this issue? please assign to me. I'll create PR.

[vadasambar]

@aiden-von I don't think anyone else is working on this. You can assign the issue to yourself with /assign @aiden-von

[vadasambar]

/assign @aiden-von

[Shubham82]

@maxpain can you confirm if this issue is resolved by PR #6235?
If yes, so we can close this issue.

[maxpain]

@Shubham82 Yes! Thanks you.

[Shubham82]

@maxpain Thanks for the confirmation.

[Shubham82]

closing this issue, as it is resolved by PR #6235

[Shubham82]

/close

[k8s-ci-robot]

@Shubham82: Closing this issue.

In response to this:

/close

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.