[
https://issues.apache.org/jira/browse/HADOOP-13249?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15321571#comment-15321571
]
Steve Loughran commented on HADOOP-13249:
-----------------------------------------
It looks like the problem is that if the thread sleeping for a retry is
interrupted, it's thrown all the way up and things expecting IOEs don't like
that.
Wrapping would fix that, but it MUST be an {{InterruptedIOException}}, so that
the fact there wasn't interrupt doesn't get lost in the chain.
However: is it that the right action? What about rethrowing the exception that
was being retried?
> RetryInvocationHandler need wrap InterruptedException in IOException when
> call Thread.sleep
> -------------------------------------------------------------------------------------------
>
> Key: HADOOP-13249
> URL: https://issues.apache.org/jira/browse/HADOOP-13249
> Project: Hadoop Common
> Issue Type: Bug
> Components: ipc
> Affects Versions: 2.8.0
> Reporter: zhihai xu
> Assignee: zhihai xu
> Attachments: HADOOP-13249.000.patch
>
>
> RetryInvocationHandler need wrap InterruptedException in IOException when
> call Thread.sleep. Otherwise InterruptedException can't be handled correctly
> by other components such as HDFS.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
