[jira] [Commented] (HADOOP-12074) in Shell.java#runCommand() while catching InterruptedException, throw new IOException(ie) instead of IOException(ie.getMessage())

[Lavkesh Lahngir (JIRA)]

    [ 
https://issues.apache.org/jira/browse/HADOOP-12074?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14577243#comment-14577243
 ] 

Lavkesh Lahngir commented on HADOOP-12074:
------------------------------------------

Apart from default constructor InterruptedIOException has a constructor which 
takes string argument. 
So no way to directly do new InterruptedIOException(cause). 
we could do.
{code}
catch(InterruptedException ie) {
      InterruptedIOException iioe = new InterruptedIOException(ie.getMessage());
      iioe.initCause(ie);
      throw iioe;
}
{code}

Or Instead of InterruptedIOException we can just throw an IOException with 
cause ie. 

{code}
catch(InterruptedException ie) {
throw newIOException(ie);
}
{code}
I feel that cause should be preserved, while wrapping exceptions. 

> in Shell.java#runCommand() while catching InterruptedException, throw new 
> IOException(ie) instead of IOException(ie.getMessage())
> ---------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-12074
>                 URL: https://issues.apache.org/jira/browse/HADOOP-12074
>             Project: Hadoop Common
>          Issue Type: Bug
>            Reporter: Lavkesh Lahngir
>            Assignee: Lavkesh Lahngir
>            Priority: Trivial
>
> Instead of creating IOException with message make InterruptedException the 
> cause of IOException, so that whoever calling shell executor can make use of 
> it. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)