On Thu, Apr 5, 2018 at 5:44 AM, Srinivas Naga Kotaru (skotaru) <
[email protected]> wrote:

> Want to validate a statement.
>
>
>
> Just assume, we have a only one node with 60 GB memory. Pod A scheduled
> with request of 10 GB and no limit. After some time, it start using 55 GB (
> ignore systems reserves for now). system now has only left with 5 GB free.
> If new podB scheduled with request of 10GB, what is systems behavior?
>
>
>
>    - Will it place Pod A and evict PodB softly since it can’t accommodate
>    both pods on same node?
>
>  Kubelet should evict pod A because it is using a higher % of memory than
it requested.

>
>    -
>    - Scheduling fail to pod B since it can’t accommodate since node
>    doesn’t enough requested memory?
>
> No, scheduling should not fail. Scheduler knows only requests and from
requests angle scheduler is fine.

>
>    -
>
>
>


For more details you might want to refer
https://github.com/kubernetes/community/blob/master/contributors/design-proposals/scheduling/pod-preemption.md#preemption---eviction-workflow


>
>
> --
>
> *Srinivas Kotaru*
>
> _______________________________________________
> dev mailing list
> [email protected]
> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
>
>
_______________________________________________
dev mailing list
[email protected]
http://lists.openshift.redhat.com/openshiftmm/listinfo/dev

Reply via email to