You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@yunikorn.apache.org by "Wu hsuang zong (Jira)" <ji...@apache.org> on 2023/09/01 14:10:00 UTC
[jira] [Created] (YUNIKORN-1951) The queues only set max resource shouldn't be eligible victim
Wu hsuang zong created YUNIKORN-1951:
----------------------------------------
Summary: The queues only set max resource shouldn't be eligible victim
Key: YUNIKORN-1951
URL: https://issues.apache.org/jira/browse/YUNIKORN-1951
Project: Apache YuniKorn
Issue Type: Bug
Components: core - scheduler
Reporter: Wu hsuang zong
The function findEligiblePreemptionVictims() is used to find preemption victims.
The code that determines the relationship between guaranteed and preemption victim is as follows:
{code:go}
guaranteed := resources.ComponentWiseMinPermissive(sq.GetActualGuaranteedResource(), sq.GetMaxResource())
if guaranteed.FitInMaxUndef(sq.GetAllocatedResource()) {
return
}
{code}
When guaranteed is not set, the maximum configuration is used. In normal circumstances, this allows the FitInMaxUndef() condition to return true, preventing the queue from becoming a victim. However, the problem arises when the queue's maximum resources are set to {memory:200, pods:0}, and yet applications with {memory:100, pods:1} can still be allocated to the queue.
Which causing FitInMaxUndef() to always return false, thus considering the queue as a victim.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@yunikorn.apache.org
For additional commands, e-mail: dev-help@yunikorn.apache.org