[
https://issues.apache.org/jira/browse/CHUKWA-196?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12703894#action_12703894
]
Jerome Boulon commented on CHUKWA-196:
--------------------------------------
Chukwa is based 4 components:
- Agent
- Collector
- Data Processor Node
- Hicc
I think that each one should have is own rpm and we should be able to install
all of them into the same machine without any conflict.
bin&conf could link to/contain files for all the installed components
$ChukwaHome/bin
$ChukwaHome/conf
Then we could have one directory per component
$ChukwaHome/Agent
$ChukwaHome/Collector
$ChukwaHome/DataProcessor
$ChukwaHome/Hicc
> Polish the deployment process for hicc
> --------------------------------------
>
> Key: CHUKWA-196
> URL: https://issues.apache.org/jira/browse/CHUKWA-196
> Project: Hadoop Chukwa
> Issue Type: Improvement
> Components: User Interface
> Environment: Redhat EL 5.1, Java 6
> Reporter: Eric Yang
>
> The deployment process for HICC is messy. The descriptors files and views
> files should be bundled as part of the chukwa tarball/rpm. The files should
> only be copied to CHUKWA_DATA_DIR, if this is a new deployment. In addition,
> remove the hard coded tomcat version from hicc.sh.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
