incubator-crunch-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rahul Sharma (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CRUNCH-4) Create profiles for Crunch builds based on Hadoop 1.0 and Hadoop 2.0
Date Fri, 20 Jul 2012 10:02:33 GMT

     [ https://issues.apache.org/jira/browse/CRUNCH-4?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Rahul Sharma updated CRUNCH-4:
------------------------------

    Attachment: 0001-Created-profiles-for-CDH3-hbase-.90.6-and-hadoop2-hb.patch

I created profiles for CDH3/hbase .90 and hadoop2/hbase .92 . Also tests/srcs were modified
for the same. By default CDH3 profile gets built. The other can be invoked by setting property
crunch.platform=2 

In hadoop 2 the Aggregate API returned output in reverse order. Also the Distributed Cache
behavior was a bit different.  
A hadoop2  run shows a bunch of deprecated warning eg:

42334 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapred.cache.files is deprecated.
Instead, use mapreduce.job.cache.files
42336 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapred.reduce.tasks is deprecated.
Instead, use mapreduce.job.reduces
42337 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapred.output.key.comparator.class
is deprecated. Instead, use mapreduce.job.output.key.comparator.class
42338 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapred.mapoutput.value.class
is deprecated. Instead, use mapreduce.map.output.value.class
42340 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapreduce.map.class is deprecated.
Instead, use mapreduce.job.map.class
42342 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapred.job.name is deprecated.
Instead, use mapreduce.job.name
42344 [Thread-288] WARN  org.apache.hadoop.conf.Configuration  - mapreduce.reduce.class is
deprecated. Instead, use mapreduce.job.reduce.class

Should we try to fix them ?

I did not made a profile for hadoop1 as a version of the same could not be found in Cloudera
repository.
                
> Create profiles for Crunch builds based on Hadoop 1.0 and Hadoop 2.0
> --------------------------------------------------------------------
>
>                 Key: CRUNCH-4
>                 URL: https://issues.apache.org/jira/browse/CRUNCH-4
>             Project: Crunch
>          Issue Type: Task
>          Components: Core
>            Reporter: Josh Wills
>            Assignee: Josh Wills
>              Labels: maven, newbie
>         Attachments: 0001-Created-profiles-for-CDH3-hbase-.90.6-and-hadoop2-hb.patch
>
>
> We need to create two Maven build profiles for Crunch, one that depends on Hadoop 1.0
(probably 1.0.3 for Hadoop and 0.90.5 for HBase) and Hadoop 2.0 (2.0.0-alpha for Hadoop and
a stable 0.92.x for HBase.)
> Reference:
> http://maven.apache.org/guides/introduction/introduction-to-profiles.html

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message