You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "Hazem Mahmoud (JIRA)" <ji...@apache.org> on 2017/06/15 21:03:00 UTC
[jira] [Commented] (MAPREDUCE-6900) Terasort replication factor
hard-coded for partition file (partFile)
[ https://issues.apache.org/jira/browse/MAPREDUCE-6900?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16051062#comment-16051062 ]
Hazem Mahmoud commented on MAPREDUCE-6900:
------------------------------------------
One idea is to use mapreduce.client.submit.file.replication, which should allow us to pass in as an argument to the terasort cmd
> Terasort replication factor hard-coded for partition file (partFile)
> --------------------------------------------------------------------
>
> Key: MAPREDUCE-6900
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6900
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Hazem Mahmoud
> Priority: Minor
>
> When running terasort on a cluster with less than 10 nodes, I get the following:
> 17/06/12 11:18:21 ERROR terasort.TeraSort: Requested replication factor of 10 exceeds maximum of 4 for /tmp/hive/tera-out/_partition.lst from 172.31.114.66
> There is no way to set this, as it is hard-coded here:
> src/main/java/org/apache/hadoop/examples/terasort/TeraInputFormat.java
> {code:java}
> DataOutputStream writer = outFs.create(partFile, true, 64*1024, (short) 10,
> outFs.getDefaultBlockSize(partFile));
> {code}
> Had to modify TeraInputFormat.java and rebuild to get it to work. This should be configurable.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: mapreduce-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: mapreduce-issues-help@hadoop.apache.org