By the way it happens on Yarn not on MRv1
each container gets 1GB at the moment.
can you try increasing memory per reducer  ? 

On Wed, Oct 31, 2012 at 9:15 PM, Eduard Skaley <> wrote:

I'm getting this Error through job execution:

16:20:26 INFO  [main]                     Job -  map 100% reduce 46%
16:20:27 INFO  [main]                     Job -  map 100% reduce 51%
16:20:29 INFO  [main]                     Job -  map 100% reduce 62%
16:20:30 INFO  [main]                     Job -  map 100% reduce 64%
16:20:32 INFO  [main]                     Job - Task Id : attempt_1351680008718_0018_r_000006_0, Status : FAILED
Error: org.apache.hadoop.mapreduce.task.reduce.Shuffle$ShuffleError: error in shuffle in fetcher#2
    at org.apache.hadoop.mapred.YarnChild$
    at Method)
    at org.apache.hadoop.mapred.YarnChild.main(
Caused by: java.lang.OutOfMemoryError: Java heap space
    at org.apache.hadoop.mapreduce.task.reduce.MapOutput.<init>(
    at org.apache.hadoop.mapreduce.task.reduce.MergeManager.unconditionalReserve(
    at org.apache.hadoop.mapreduce.task.reduce.MergeManager.reserve(
    at org.apache.hadoop.mapreduce.task.reduce.Fetcher.copyMapOutput(
    at org.apache.hadoop.mapreduce.task.reduce.Fetcher.copyFromHost(

16:20:33 INFO  [main]                     Job -  map 100% reduce 65%
16:20:36 INFO  [main]                     Job -  map 100% reduce 67%
16:20:39 INFO  [main]                     Job -  map 100% reduce 69%
16:20:41 INFO  [main]                     Job -  map 100% reduce 70%
16:20:43 INFO  [main]                     Job -  map 100% reduce 71%

I have no clue what the issue could be for this. I googled this issue and checked several sources of possible solutions but nothing does fit.

I saw this jira entry which could fit:

Here somebody recommends to increase the value for the property dfs.datanode.max.xcievers / dfs.datanode.max.receiver.threads to 4096, but this is the value for our cluster.

The issue with the to small input files doesn't fit I think, because the map phase reads 137 files with each 130MB. Block Size is 128MB.

The cluster uses version
2.0.0-cdh4.1.1, 581959ba23e4af85afd8db98b7687662fe9c5f20.


Nitin Pawar