infra-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Meghna Baijal (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (INFRA-15071) MXNet: Jenkins Job/Slave does not timeout even after it fails
Date Tue, 26 Sep 2017 01:03:00 GMT

     [ https://issues.apache.org/jira/browse/INFRA-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Meghna Baijal updated INFRA-15071:
----------------------------------
    Status: Waiting for Infra  (was: Waiting for user)

> MXNet: Jenkins Job/Slave does not timeout even after it fails
> -------------------------------------------------------------
>
>                 Key: INFRA-15071
>                 URL: https://issues.apache.org/jira/browse/INFRA-15071
>             Project: Infrastructure
>          Issue Type: Project
>          Components: Jenkins
>            Reporter: Meghna Baijal
>
> Many of the recent (past 1 week) MXNet build jobs have been running for over 14 hours.
Most of them fail in connecting to some slave. However, neither the job nor the slave times
out and the jobs just stays stuck/hangs. 
> For example this is happening in this job which ran for 13 hours - 
> https://builds.apache.org/blue/organizations/jenkins/incubator-mxnet/detail/master/329/pipeline/49
> Are jenkinsfile is configured to timeout each stage after a maximum of 2 hours. I also
believe that a jenkins slave is supposed to timeout after 30 minutes. 
> Can someone please provide some insight?



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

Mime
View raw message