[
https://issues.apache.org/jira/browse/HDFS-14258?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
BELUGA BEHR updated HDFS-14258:
-------------------------------
Status: Patch Available (was: Open)
[~elgoiri] Putting up a new patch with the requested changes.
Setting the wait time to a large number or zero still changes the behavior.
Right now, if the reconfiguration happens, it is "fire and forget."
With the use of a {{Semaphore}} it's "all or nothing", it's just a matter of
how long to wait for "all" to occur. Setting to a large number blocks until it
completes (maybe this is the direction to go, but then there's always the risk
of a hang if some number of the block mover threads are stuck; an unlikely, but
possible scenario). Setting it to zero means try, but if it's not able to
immediately reduce the pool of workers, it fails. Also a possible solution.
Let me know.
> Introduce Java Concurrent Package To DataXceiverServer Class
> ------------------------------------------------------------
>
> Key: HDFS-14258
> URL: https://issues.apache.org/jira/browse/HDFS-14258
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Affects Versions: 3.2.0
> Reporter: BELUGA BEHR
> Assignee: BELUGA BEHR
> Priority: Minor
> Attachments: HDFS-14258.1.patch, HDFS-14258.2.patch,
> HDFS-14258.3.patch, HDFS-14258.4.patch, HDFS-14258.5.patch, HDFS-14258.6.patch
>
>
> * Use Java concurrent package to replace current facilities in
> {{DataXceiverServer}}.
> * A little bit of extra clean up
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
