You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@yunikorn.apache.org by "Weiwei Yang (Jira)" <ji...@apache.org> on 2021/09/17 16:01:00 UTC
[jira] [Resolved] (YUNIKORN-848) Scheduler occasionally crashed
due to concurrent r/w map error
[ https://issues.apache.org/jira/browse/YUNIKORN-848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Weiwei Yang resolved YUNIKORN-848.
----------------------------------
Resolution: Fixed
> Scheduler occasionally crashed due to concurrent r/w map error
> ----------------------------------------------------------------
>
> Key: YUNIKORN-848
> URL: https://issues.apache.org/jira/browse/YUNIKORN-848
> Project: Apache YuniKorn
> Issue Type: Bug
> Components: shim - kubernetes
> Reporter: Weiwei Yang
> Assignee: Weiwei Yang
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.0.0
>
>
> It is found that occasionally the scheduler could crash due to concurrent map r/w error. The root cause is because during the evaluation of the predicates, we pass a nodesInfo map as a reference to the K8s side, and when concurrently there is a node being add/remove to the cluster, it could possibly cause the concurrent read/write issue.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@yunikorn.apache.org
For additional commands, e-mail: issues-help@yunikorn.apache.org