[
https://issues.apache.org/jira/browse/CRUNCH-140?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14940017#comment-14940017
]
Landon Robinson commented on CRUNCH-140:
----------------------------------------
Hi Micah! I actually just figured it out: was an issue with the data nodes not
having the hadoop client installed on them, so when we gave them a input path,
they were checking locally (local path), not HDFS. Our solution was to change
out passed path parameters from /path/in/hdfs/ to hdfs://namenode/path/in/hdfs/.
that worked, and now we can run Crunch jar in oozie!
> Input source Text does not exist!
> ----------------------------------
>
> Key: CRUNCH-140
> URL: https://issues.apache.org/jira/browse/CRUNCH-140
> Project: Crunch
> Issue Type: Bug
> Environment: Ubuntu 12.04
> Hadoop 0.20.205
> Apache Crunch 0.4.0-incubating (binary) tar.gz
> Reporter: DeepakVohra
>
> The Crunch examples org.apache.crunch.examples.WordCount,
> org.apache.crunch.examples.AverageBytesByIP,
> org.apache.crunch.examples.TotalBytesByIP generate error:
> Input source Text (the sample input) does not exist!
> Input path does not include "globs/wildcards" as noted in another issue.
> https://issues.apache.org/jira/browse/CRUNCH-131
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
