[
https://issues.apache.org/jira/browse/HDFS-15659?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17222090#comment-17222090
]
Ayush Saxena commented on HDFS-15659:
-------------------------------------
[~aajisaka] IIRC couple of days back you had objections having this config
disabled in a test. I didn't follow up, and I lost the track of that Jira too.
Is that concern not there now?
I think there the number of datanodes were increased instead? If we are doing
this, I don't think that fix would be required? So, If this concludes, We can
revert that. Increasing datanodes has adverse affect on test performance,
though minor
> Set dfs.namenode.redundancy.considerLoad to false in MiniDFSCluster
> -------------------------------------------------------------------
>
> Key: HDFS-15659
> URL: https://issues.apache.org/jira/browse/HDFS-15659
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: test
> Reporter: Akira Ajisaka
> Priority: Major
>
> dfs.namenode.redundancy.considerLoad is true by default and it is causing
> many test failures. Let's disable it in MiniDFSCluster.
> Originally reported by [~weichiu]:
> https://github.com/apache/hadoop/pull/2410#pullrequestreview-516144442
> {quote}
> i've certain seen this option causing test failures in the past.
> Maybe we should turn it off by default in MiniDDFSCluster, and only enable it
> for specific tests.
> {quote}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
