infra-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Allen Wittenauer (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (INFRA-15424) Jenkins stops triggering/ scheduling builds
Date Thu, 02 Nov 2017 05:48:01 GMT

    [ https://issues.apache.org/jira/browse/INFRA-15424?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16235216#comment-16235216
] 

Allen Wittenauer commented on INFRA-15424:
------------------------------------------

FWIW, when the scheduling bug hits, it seems to impact every job off of precommit-admin. 
This includes Precommit-Yetus-Build, which has a pretty wide net cast.

As a point of interest, the vast majority of Hadoop builds are now running with a greatly
restricted memory (docker -m 20g) and ulimit (5k).  This includes a set of branches that I
found were running amok.  Even after I deployed the fixes, things were still unstable on the
nodes though. :(  [See INFRA-15373 and associated JIRA issues for more info.]

If there's anything I can do to help, let me know. 

> Jenkins stops triggering/ scheduling builds
> -------------------------------------------
>
>                 Key: INFRA-15424
>                 URL: https://issues.apache.org/jira/browse/INFRA-15424
>             Project: Infrastructure
>          Issue Type: Project
>          Components: Jenkins
>            Reporter: Daniel Takamori
>            Assignee: Daniel Takamori
>            Priority: Major
>
> We've seen multiple cases of Jenkins stopping scheduling of jobs while there are available
nodes matching labels.  
> Things to look at:
> Logs
> Plugin changes
> Regressions
> Tangentially, Blue Ocean might be causing increased load on the Jenkins server due to
'auto refreshing' 



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message