[jira] [Commented] (HDFS-17765) Provide a mechanism for reordering datanode priority when reading

["Istvan Toth (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HDFS-17765?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17939549#comment-17939549 ] 

Istvan Toth commented on HDFS-17765:
------------------------------------

This ticket kind of duplicates HDFS-4754 but leaves open the possibility for other solutions.

> Provide a mechanism for reordering datanode priority  when reading
> ------------------------------------------------------------------
>
>                 Key: HDFS-17765
>                 URL: https://issues.apache.org/jira/browse/HDFS-17765
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs-client
>            Reporter: Istvan Toth
>            Priority: Major
>
> HBASE-6435 has added an ugly hack to avoid long waits for HDFS timeouts when reading data whoose primary DataNode is known to be down.
> With JEP-416 that solution is no longer working.
> Provide some solution for HBASE-6435 that does require horrible reflection based hacks.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org