[
https://issues.apache.org/jira/browse/HDFS-11079?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15630303#comment-15630303
]
Mingliang Liu edited comment on HDFS-11079 at 11/3/16 6:33 AM:
---------------------------------------------------------------
You're correct [~arpitagarwal] I just had a look at the code and found a few
places in the unit tests that restart the NN, checkpoint, and etc. It's very
likely the cluster is left with a bad state.
Let's simply keep the individual cluster as it is. -That's said, we don't need
the {{refreshClusterState();}} after we build a brand new cluster in
{{setUp()}}.-
was (Author: liuml07):
You're correct [~arpitagarwal] I just had a look at the code and found a few
places in the unit tests that restart the NN, checkpoint, and etc. It's very
likely the cluster is left with a bad state.
Let's simply keep the individual cluster as it is. That's said, we don't need
the {{refreshClusterState();}} after we build a brand new cluster in
{{setUp()}}.
> TestQuotaByStorageType#setUp does not need refresh the cluster
> --------------------------------------------------------------
>
> Key: HDFS-11079
> URL: https://issues.apache.org/jira/browse/HDFS-11079
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: test
> Reporter: Xiaobing Zhou
> Assignee: Xiaobing Zhou
> Priority: Minor
> Attachments: HDFS-11079.000.patch
>
>
> -When tests are added to TestQuotaByStorageType by HDFS-11065, it's been
> noticed that it takes minutes to launch MiniDFSClusters for every tests. This
> tries to refactor the code to use a static shared MiniDFSCluster to
> accelerate tests.-
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
