[
https://issues.apache.org/jira/browse/HDFS-11182?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15740497#comment-15740497
]
Arpit Agarwal commented on HDFS-11182:
--------------------------------------
Hi [~manju_hadoop],
bq. The invokeCallback releases permit in Semaphore and then does a tryAcquire
which would allow it to aquire the permit it just released and hence would go
and call the callback.call method which would release the latch by the
countDown.
Since the number of permits is initially negative, it will take
references.size() releases before any acquire call succeeds. So the first
references.size() -1 threads will not be able to acquire the permit they just
released.
> Update DataNode to use DatasetVolumeChecker
> -------------------------------------------
>
> Key: HDFS-11182
> URL: https://issues.apache.org/jira/browse/HDFS-11182
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode
> Reporter: Arpit Agarwal
> Assignee: Arpit Agarwal
>
> Update DataNode to use the DatasetVolumeChecker class introduced in
> HDFS-11149 to parallelize disk checks.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
