accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Hsieh (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (ACCUMULO-1550) Create ACCUMULO_CONF_DIR environment variable
Date Mon, 15 Jul 2013 22:18:49 GMT

     [ https://issues.apache.org/jira/browse/ACCUMULO-1550?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jonathan Hsieh updated ACCUMULO-1550:
-------------------------------------

    Attachment: accumulo-1550-1.4.patch
                accumulo-1550-trunk.patch

All versions pass unit tests.

The trunk version applies cleanly to 1.5 and to trunk.  

I've tested the 1.5 version by deploying sucessfully a using start-all.sh to start a 1.5 cluster
using the ACCUMULO_CONF_DIR env variable where the accumulo/conf dir was renamed and not present.
 It started.

I've been testing more rigorously with the 1.4 patch and have been running continuous ingets
and randomwalks with a similar use the ACCUMULO_CONF_DIR env variable and rename the accumulo/conf.


There are a few places we actually want to use ACCUMULO_HOME/conf like when we copy example
dirs in testChangingDirectory, and in config.sh where we default to ACCUMULO_HOME/conf id
no ACCUMULO_CONF_DIR is specified.
                
> Create ACCUMULO_CONF_DIR environment variable
> ---------------------------------------------
>
>                 Key: ACCUMULO-1550
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-1550
>             Project: Accumulo
>          Issue Type: Improvement
>          Components: scripts
>    Affects Versions: 1.4.3, 1.5.0
>            Reporter: Jonathan Hsieh
>            Assignee: Jonathan Hsieh
>             Fix For: 1.6.0
>
>         Attachments: accumulo-1550-1.4.patch, accumulo-1550-trunk.patch
>
>
> Hadoop (and many of the other projects in the hadoop stack) have a HADOOP_CONF_DIR (<project>_CONF_DIR)
environment variable that is used to start daemons using the configuration specified.  
> This is convenient because it decouples the configuration from the distributions -- daemons
can be run with different configs without having to perform stateful file operations.  
> For example one could run accumulo like this:
> {code}
> ACCUMULO_CONF_DIR=/etc/accumulo/conf accumulo ....
> {code}
> or set env variables in .bashrc/jenkins scripts for different configurations
> {code}
> export ACCUMULO_CONF_DIR=$ACCUMULO_HOME/conf/examples/1GB/standalone
> .. run test
> export ACCUMULO_CONF_DIR=$ACCUMULO_HOME/conf/examples/1GB/native-standalone
> .. run test
> {code}

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message