[
https://issues.apache.org/jira/browse/HADOOP-11984?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14549834#comment-14549834
]
Chris Nauroth commented on HADOOP-11984:
----------------------------------------
bq. MAVEN_OPTS is set to "-Xmx3072m -XX:MaxPermSize=768m"
Thanks for the tip. That could be a factor. In my environment, I set
{{MAVEN_OPTS}} with a max 1 GB heap. The forked JUnit processes are set to use
1 GB max heap each. (This is coded into the pom.xml, so this part won't vary
between build environments.) That means a pre-commit runs on Jenkins would be
3 GB + 4 * 1 GB = 7 GB total. If a significant amount of the host's RAM is
already allocated to Jenkins and whatever else runs on the host, then maybe
we're resource-constrained. I'll ask infra for verification.
> Enable parallel JUnit tests in pre-commit.
> ------------------------------------------
>
> Key: HADOOP-11984
> URL: https://issues.apache.org/jira/browse/HADOOP-11984
> Project: Hadoop Common
> Issue Type: Improvement
> Components: scripts
> Reporter: Chris Nauroth
> Assignee: Chris Nauroth
> Attachments: HADOOP-11984.001.patch, HADOOP-11984.002.patch,
> HADOOP-11984.003.patch, HADOOP-11984.004.patch, HADOOP-11984.005.patch
>
>
> HADOOP-9287 and related issues implemented the parallel-tests Maven profile
> for running JUnit tests in multiple concurrent processes. This issue
> proposes to activate that profile during pre-commit to speed up execution.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
