[
https://issues.apache.org/jira/browse/HDFS-17000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17721297#comment-17721297
]
Ayush Saxena commented on HDFS-17000:
-------------------------------------
yep, we should change int i to long.
Though technically it won't ever go that far, the code with go into the catch
block post 1024 * 1024 * 5, but makes sense to change to long.
[~Marcono1234] can you raise a PR to change to long. Let me know if you face
any issues. can try help
> Potential infinite loop in
> TestDFSStripedOutputStreamUpdatePipeline.testDFSStripedOutputStreamUpdatePipeline
> ------------------------------------------------------------------------------------------------------------
>
> Key: HDFS-17000
> URL: https://issues.apache.org/jira/browse/HDFS-17000
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Marcono1234
> Priority: Major
>
> The method
> {{TestDFSStripedOutputStreamUpdatePipeline.testDFSStripedOutputStreamUpdatePipeline}}
> contains the following line:
> {code}
> for (int i = 0; i < Long.MAX_VALUE; i++) {
> {code}
> [GitHub source
> link|https://github.com/apache/hadoop/blob/4ee92efb73a90ae7f909e96de242d216ad6878b2/hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestDFSStripedOutputStreamUpdatePipeline.java#L48]
> Because {{i}} is an {{int}} the condition {{i < Long.MAX_VALUE}} will always
> be true and {{i}} will simply overflow.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]