[
https://issues.apache.org/jira/browse/YUNIKORN-796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17398384#comment-17398384
]
Chaoran Yu commented on YUNIKORN-796:
-------------------------------------
Thanks [~wilfreds]. We've considered these factors but they don't explain what
we observed. We are using fair sorting. And our pods don't bind to a volume
(they do use emptyDir volumes though). I checked that the pods don't have
special settings like affinity or anti-affinity. I don't have the logs now but
I'll follow up here when I get some
> Pod scheduling could disproportionately concentrate on one node
> ---------------------------------------------------------------
>
> Key: YUNIKORN-796
> URL: https://issues.apache.org/jira/browse/YUNIKORN-796
> Project: Apache YuniKorn
> Issue Type: Bug
> Components: core - scheduler
> Reporter: Chaoran Yu
> Priority: Major
>
> We've observed a few cases when YK keeps scheduling incoming pods to just one
> node, even though the cluster has many other nodes that have abundant
> resources. Not sure what could be causing this behavior because we don't see
> it in every K8s cluster we have. Maybe somewhere in the node sorting
> algorithm, a node could be favored over all other nodes under some conditions?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
