[
https://issues.apache.org/jira/browse/MAPREDUCE-6724?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15389986#comment-15389986
]
Gera Shegalov commented on MAPREDUCE-6724:
------------------------------------------
[~haibochen], what I mean if you revert your change to
hadoop-mapreduce-project/hadoop-mapreduce-client/hadoop-mapreduce-client-core/src/test/java/org/apache/hadoop/mapreduce/task/reduce/TestMergeManager.java
, and apply my shorter
[change|https://issues.apache.org/jira/browse/MAPREDUCE-6724?focusedCommentId=15384827&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15384827]
to it, I think it tests the bugs you are fixing: it will fail without the fix,
and succeed with the fix.
> Unsafe conversion from long to int in MergeManagerImpl.unconditionalReserve()
> -----------------------------------------------------------------------------
>
> Key: MAPREDUCE-6724
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6724
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: mrv2
> Reporter: Haibo Chen
> Assignee: Haibo Chen
> Attachments: mapreduce6724.001.patch, mapreduce6724.002.patch,
> mapreduce6724.003.patch, mapreduce6724.004.patch, mapreduce6724.005.patch,
> mapreduce6724.006.patch
>
>
> When shuffle is done in memory, MergeManagerImpl converts the requested size
> to an int to allocate an instance of InMemoryMapOutput. This results in an
> overflow if the requested size is bigger than Integer.MAX_VALUE and
> eventually causes the reducer to fail.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
