[
https://issues.apache.org/jira/browse/YUNIKORN-796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17398868#comment-17398868
]
Weiwei Yang edited comment on YUNIKORN-796 at 8/13/21, 7:02 PM:
----------------------------------------------------------------
I think there are a few things that need to be checked
# pod affinity/anti-affinity
# node affinity/anti-affinity
# node used resources in YK's view (via REST API)
# is reservation enabled or disabled? check logs and see if there are pods got
reserved on some nodes
[~yuchaoran2011] pls let us know if you have more findings
was (Author: wwei):
I think there are a few things that need to be checked
# pod affinity/anti-affinity
# node affinity/anti-affinity
# node used resources in YK's view (via REST API)
# is reservation enabled or disabled? check logs and see if there are pods got
reserved on some nodes
[~yuchaoran2011] pls let us know if you have more findings
> 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]
