[
https://issues.apache.org/jira/browse/HDFS-4018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13471677#comment-13471677
]
Aaron T. Myers commented on HDFS-4018:
--------------------------------------
The patch looks pretty good to me. A few little comments:
# This will make a test failure harder to debug, not easier, since we'll be
losing the stack trace. We should just let the IOE get thrown:
{code}
+ } catch (IOException e) {
+ Assert.fail("Failed to add NN to cluster");
{code}
# Not sure why you made this change. This is in general good practice so that
we get the actual stack trace of why the minicluster failed to start, instead
of an NPE:
{code}
- if(cluster != null)
- cluster.shutdown();
+ cluster.shutdown();
{code}
> TestDataNodeMultipleRegistrations#testMiniDFSClusterWithMultipleNN is missing
> some cluster cleanup
> --------------------------------------------------------------------------------------------------
>
> Key: HDFS-4018
> URL: https://issues.apache.org/jira/browse/HDFS-4018
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 2.0.0-alpha
> Reporter: Eli Collins
> Assignee: Eli Collins
> Priority: Minor
> Attachments: hdfs-4018.txt
>
>
> TestDataNodeMultipleRegistrations#testMiniDFSClusterWithMultipleNN does not
> unconditionally shutdown the cluster if a call to addNameNode fails, so if
> one of these calls fails other tests may failed due to the NN directory
> already being locked.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
