[
https://issues.apache.org/jira/browse/HDFS-16126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17380106#comment-17380106
]
Renukaprasad C commented on HDFS-16126:
---------------------------------------
[~lei w] Thanks for reporting the issue.
org.apache.hadoop.hdfs.server.datanode.DiskBalancer.VolumePair#hashCode
org.apache.hadoop.hdfs.server.datanode.DiskBalancer.VolumePair#equals
These 2 methods already implemented in VolumePair. Anything missing out of this?
> VolumePair should override hashcode() method
> -----------------------------------------------
>
> Key: HDFS-16126
> URL: https://issues.apache.org/jira/browse/HDFS-16126
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: diskbalancer
> Reporter: lei w
> Priority: Minor
>
> Now we use a map to check one plan with more than one line of same
> VolumePair in createWorkPlan(final VolumePair volumePair, Step step) , code
> is as flow:
> {code:java}
> private void createWorkPlan(final VolumePair volumePair, Step step)
> throws DiskBalancerException {
> // ...
> // In case we have a plan with more than
> // one line of same VolumePair
> // we compress that into one work order.
> if (workMap.containsKey(volumePair)) { // To check use map
> bytesToMove += workMap.get(volumePair).getBytesToCopy();
> }
> // ...
> }
> {code}
> I found the object volumePair is always a new object and without hashcode()
> method, So use a map to check is invalid. Should we add hashcode() in
> VolumePair ?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
