[
https://issues.apache.org/jira/browse/HDFS-1170?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12870048#action_12870048
]
Steve Loughran commented on HDFS-1170:
--------------------------------------
And catch exceptions in child threads, relay them up, have a better way of
stopping the worker threads than just interrupt(), which isn't checked on every
operation.
> Add more assertions to TestLargeDirectoryDelete
> -----------------------------------------------
>
> Key: HDFS-1170
> URL: https://issues.apache.org/jira/browse/HDFS-1170
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: test
> Affects Versions: 0.22.0
> Reporter: Steve Loughran
> Assignee: Steve Loughran
> Priority: Minor
>
> This issue is for a patch which will only add some more asserts to the
> TestLargeDirectoryDelete test case, to help identify the causes of HDFS-615
> and HDFS-816.
> Proposed
> # split the chained operations into one per line for better stack trace
> diagnostics
> # add {{assertNotNull}} assertions for better diagnostics.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
