[
https://issues.apache.org/jira/browse/HADOOP-3513?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12603639#action_12603639
]
Konstantin Shvachko commented on HADOOP-3513:
---------------------------------------------
> Wouldn't it better to wait until the namenode leaves safemode rather than
> forcing it out?
I you start the name-node with an existing image and all data-nodes are down,
which is what the benchmark does,
then there is no chance the name-node will leave safe-mode naturally. So I have
to force it out.
> Improve NNThroughputBenchmark log messages.
> -------------------------------------------
>
> Key: HADOOP-3513
> URL: https://issues.apache.org/jira/browse/HADOOP-3513
> Project: Hadoop Core
> Issue Type: Bug
> Components: test
> Affects Versions: 0.17.0
> Reporter: Konstantin Shvachko
> Assignee: Konstantin Shvachko
> Fix For: 0.18.0
>
> Attachments: NNThroughputBenchmarkLogMsg.patch
>
>
> The block report benchmark should not print return values of heartbeats.
> It should also leave safe mode before creating files.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
