[
https://issues.apache.org/jira/browse/MAPREDUCE-6234?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14308900#comment-14308900
]
Varun Saxena commented on MAPREDUCE-6234:
-----------------------------------------
[~ozawa],
Below is the description of the config.
{code}
<description>
The amount of memory to request from the scheduler for each
map task. If this is not specified or is non-positive, it is inferred from
mapreduce.map.java.opts and mapreduce.job.heap.memory-mb.ratio.
If java-opts are also not specified, we set it to 1024.
</description>
{code}
-1 basically indicates map memory has to be taken from java opts first. And if
not found, will be set as 1024 by default. So effective default value is
1024(if java opts don't exist). This is mentioned in config description.
I don't think not making DEFAULT_MAP_MEMORY_MB as -1 is an oversight. That is
how the developer has written the code. It had an impact on {{TestJobConf}} in
hadoop-mapreduce-jobclient though. MAPREDUCE-6239 is raised to address
consolidating of two {{TestJobConf}} classes.
Maybe MAPREDUCE-6223 can fix the test case failure and this can address making
it consistent.
Thoughts ?
> MRJobConfig.DEFAULT_*_MEMORY_MB should be consistent with mapred-default.xml
> ----------------------------------------------------------------------------
>
> Key: MAPREDUCE-6234
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6234
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: contrib/gridmix, mrv2
> Reporter: Masatake Iwasaki
> Assignee: Masatake Iwasaki
> Attachments: MAPREDUCE-6234.001.patch, MAPREDUCE-6234.002.patch
>
>
> TestHighRamJob fails by this.
> {code}
> -------------------------------------------------------
> T E S T S
> -------------------------------------------------------
> Running org.apache.hadoop.mapred.gridmix.TestHighRamJob
> Tests run: 1, Failures: 1, Errors: 0, Skipped: 0, Time elapsed: 1.162 sec <<<
> FAILURE! - in org.apache.hadoop.mapred.gridmix.TestHighRamJob
> testHighRamFeatureEmulation(org.apache.hadoop.mapred.gridmix.TestHighRamJob)
> Time elapsed: 1.102 sec <<< FAILURE!
> java.lang.AssertionError: expected:<1024> but was:<-1>
> at org.junit.Assert.fail(Assert.java:88)
> at org.junit.Assert.failNotEquals(Assert.java:743)
> at org.junit.Assert.assertEquals(Assert.java:118)
> at org.junit.Assert.assertEquals(Assert.java:555)
> at org.junit.Assert.assertEquals(Assert.java:542)
> at
> org.apache.hadoop.mapred.gridmix.TestHighRamJob.testHighRamConfig(TestHighRamJob.java:98)
> at
> org.apache.hadoop.mapred.gridmix.TestHighRamJob.testHighRamFeatureEmulation(TestHighRamJob.java:117)
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
