[
https://issues.apache.org/jira/browse/HDFS-14769?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16914505#comment-16914505
]
Íñigo Goiri commented on HDFS-14769:
------------------------------------
We should avoid moving all the imports.
BTW, I think I have seen some similar JIRA.
Actually, it looks like mine: HDFS-12904
> Add throttler for DataNode#DataTransfer
> ---------------------------------------
>
> Key: HDFS-14769
> URL: https://issues.apache.org/jira/browse/HDFS-14769
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Lisheng Sun
> Assignee: Lisheng Sun
> Priority: Major
> Attachments: HDFS-14769.001.patch
>
>
> When a amount of block replicas are moving, the bandwidth will be full
> without throtter in DataNode#DataTransfer.
> This will affect normal reading and writing.
> {code:java}
> /**
> * Do the deed, write the bytes
> */
> @Override
> public void run() {
> // send data & checksum
> blockSender.sendBlock(out, unbufOut, null);
> }{code}
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
