[
https://issues.apache.org/jira/browse/HADOOP-11475?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14349195#comment-14349195
]
Sangjin Lee commented on HADOOP-11475:
--------------------------------------
Sorry [~ted_yu], I was out for a few days. I'm +1 with the solution.
> Utilize try-with-resource to close StopWatch
> --------------------------------------------
>
> Key: HADOOP-11475
> URL: https://issues.apache.org/jira/browse/HADOOP-11475
> Project: Hadoop Common
> Issue Type: Improvement
> Reporter: Ted Yu
> Assignee: Ted Yu
> Priority: Minor
> Attachments: hadoop-11475-001.patch
>
>
> Currently the stop() method of StopWatch is called without using finally
> clause.
> This can result in resource leak if there is IOE thrown.
> Here is one example from Journal#journal():
> {code}
> StopWatch sw = new StopWatch();
> sw.start();
> curSegment.flush(shouldFsync);
> sw.stop();
> {code}
> If curSegment.flush() throws IOE, sw would be left unclosed.
> Propose using try-with-resource structure to close the StopWatch.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
