You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@phoenix.apache.org by "Geoffrey Jacoby (JIRA)" <ji...@apache.org> on 2018/10/03 17:25:00 UTC
[jira] [Updated] (PHOENIX-4947) PhoenixMapReduceUtil should add all
required classes to Job
[ https://issues.apache.org/jira/browse/PHOENIX-4947?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Geoffrey Jacoby updated PHOENIX-4947:
-------------------------------------
Description:
MapReduce jobs need to know which jars to send to the mapper and reducer nodes. Phoenix requires quite a few classes that don't get shipped by default. If a developer using the Phoenix MapReduce integration forgets one, the job will work correctly when run on a single node, such as a developer workstation, but break when deployed to a distributed cluster.
Getting this right requires either trial and error, or deep knowledge of Phoenix internals. We should transparently do the right thing when someone calls PhoenixMapReduceUtil.setInput and setOutput.
> PhoenixMapReduceUtil should add all required classes to Job
> -----------------------------------------------------------
>
> Key: PHOENIX-4947
> URL: https://issues.apache.org/jira/browse/PHOENIX-4947
> Project: Phoenix
> Issue Type: Improvement
> Reporter: Geoffrey Jacoby
> Priority: Major
>
> MapReduce jobs need to know which jars to send to the mapper and reducer nodes. Phoenix requires quite a few classes that don't get shipped by default. If a developer using the Phoenix MapReduce integration forgets one, the job will work correctly when run on a single node, such as a developer workstation, but break when deployed to a distributed cluster.
> Getting this right requires either trial and error, or deep knowledge of Phoenix internals. We should transparently do the right thing when someone calls PhoenixMapReduceUtil.setInput and setOutput.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)