[
https://issues.apache.org/jira/browse/HDFS-9139?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15133596#comment-15133596
]
Mingliang Liu commented on HDFS-9139:
-------------------------------------
{code:title=org/apache/hadoop/hdfs/server/datanode/TestDataNodeMetrics.java}
283 MetricsRecordBuilder rbNew =
getMetrics(datanode.getMetrics().name());
284 long endWriteValue = getLongCounter("TotalWriteTime", rbNew);
285 long endReadValue = getLongCounter("TotalReadTime", rbNew);
286 // Lets Metric system update latest metrics
287 Thread.sleep(100);
288 assertTrue(endReadValue > startReadValue);
288 assertTrue(endWriteValue > startWriteValue);
{code}
Should we call the sleep *before* getting the end counter?
Hoping to resolve recent build failure reported by [HDFS-9766].
> Enable parallel JUnit tests for HDFS Pre-commit
> ------------------------------------------------
>
> Key: HDFS-9139
> URL: https://issues.apache.org/jira/browse/HDFS-9139
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: test
> Reporter: Vinayakumar B
> Assignee: Vinayakumar B
> Fix For: 2.8.0
>
> Attachments: HDFS-9139.01.patch, HDFS-9139.02.patch,
> HDFS-9139.03.patch, HDFS-9139.04.patch
>
>
> Forked from HADOOP-11984,
> With the initial and significant work from [~cnauroth], this Jira is to track
> and support parallel tests' run for HDFS Precommit
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
