[
https://issues.apache.org/jira/browse/HIVE-30260?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18144320#comment-18144320
]
Stamatis Zampetakis commented on HIVE-30260:
--------------------------------------------
I checked the Jenkins logs and there are many SEVERE failures around queue and
node provisioning:
{noformat}
2026-11-13 08:57:07.189+0000 [id=51] SEVERE
hudson.triggers.SafeTimerTask#run: Timer task
hudson.model.Queue$MaintainTask@674dbc89 failed
java.lang.NullPointerException
2026-11-13 08:57:12.190+0000 [id=49] SEVERE
hudson.triggers.SafeTimerTask#run: Timer task
hudson.model.Queue$MaintainTask@674dbc89 failed
java.lang.NullPointerException
2026-11-13 08:57:16.430+0000 [id=42] SEVERE
hudson.triggers.SafeTimerTask#run: Timer task
hudson.model.LoadStatistics$LoadStatisticsUpdater@6651acff failed
java.lang.NullPointerException
2026-11-13 08:57:16.812+0000 [id=43] SEVERE
hudson.triggers.SafeTimerTask#run: Timer task
hudson.slaves.NodeProvisioner$NodeProvisionerInvoker@2200fcc1 failed
java.lang.NullPointerException
2026-11-13 08:57:17.191+0000 [id=51] SEVERE
hudson.triggers.SafeTimerTask#run: Timer task
hudson.model.Queue$MaintainTask@674dbc89 failed
java.lang.NullPointerException
{noformat}
Unfortunately the full stack trace of the NPE is not present in the logs very
likely due to {{-XX:+OmitStackTraceInFastThrow}} enabled by default.
I restarted Jenkins by deleting the respective pod (kubectl delete pod
jenkins-558bfdb466-86pnp) and after the restart PRs are advancing again and
there are no SEVERE errors in the logs.
Very likely the problem has something to do with the autoscaler configuration
in GCP and the Kubernetes plugin that manages the provisioning but at the
moment don't have time to dig further.
> Jenkins jobs are hitting timeouts and getting aborted
> -----------------------------------------------------
>
> Key: HIVE-30260
> URL: https://issues.apache.org/jira/browse/HIVE-30260
> Project: Hive
> Issue Type: Bug
> Components: Testing Infrastructure
> Reporter: Stamatis Zampetakis
> Assignee: Stamatis Zampetakis
> Priority: Blocker
>
> All precommit jobs are getting aborted since yesterday (November 12, 2026)
> due to timeout limits.
> Example:
> https://ci.hive.apache.org/job/hive-precommit/view/change-requests/job/PR-6962/1/console
> {noformat}
> 19:54:20 Still waiting to schedule task
> 07:54:05 Cancelling nested steps due to timeout
> [Pipeline] // node
> [Pipeline] }
> [Pipeline] // timeout
> [Pipeline] }
> [Pipeline] // podTemplate
> [Pipeline] }
> [Pipeline] // timestamps
> [Pipeline] }
> Lock released on resource [Label: hive-precommit, Quantity: 1]
> [Pipeline] // lock
> [Pipeline] echo
> [tests failed]
> [Pipeline] End of Pipeline
> Timeout has been exceeded
> org.jenkinsci.plugins.workflow.actions.ErrorAction$ErrorId:
> f21f0910-de64-4e31-9315-99f857a364db
> {noformat}
> The same pattern appears on all latest pull requests.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)