[
https://issues.apache.org/jira/browse/MAPREDUCE-5452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13852412#comment-13852412
]
Vinod Kumar Vavilapalli commented on MAPREDUCE-5452:
----------------------------------------------------
I think it is a case of using the wrong constructor. Ideally applications
should never construct or parse TaskID strings, but where that is not possible,
one should use the appropriate constructors which provide all the information
or the forName API. For e.g, even if we fix this, toString will print the wrong
ID (the default).
I just saw HIVE-4216, the comment by Andrey
[here|https://issues.apache.org/jira/browse/HIVE-4216?focusedCommentId=13848792&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13848792]
seems like a reasonable solution.
There are far too many empty constructors like this in all of Hadoop that
aren't expected to be used directly. Fixing them is a lot of work for too
little ROI. I propose that we close it as WON'T FIX.
> NPE in TaskID toString when default constructor is used
> -------------------------------------------------------
>
> Key: MAPREDUCE-5452
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-5452
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Affects Versions: 2.0.5-alpha
> Reporter: Brock Noland
> Priority: Minor
>
> If you call TaskID.toString() after using the default the constructor
> toString() NPE's because taskType is null.
--
This message was sent by Atlassian JIRA
(v6.1.4#6159)
