You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@bigtop.apache.org by "Guo Ruijing (JIRA)" <ji...@apache.org> on 2014/07/10 14:41:04 UTC
[jira] [Comment Edited] (BIGTOP-1339) hadoop pseudo configuration
can be default configuration
[ https://issues.apache.org/jira/browse/BIGTOP-1339?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14057180#comment-14057180 ]
Guo Ruijing edited comment on BIGTOP-1339 at 7/10/14 12:40 PM:
---------------------------------------------------------------
Hi, Mark,
I agree with you that /etc/<component>/conf is a symlink and can be changed via update-alternative in %post. Maybe I didn't explain very well and pls ignore the above comments.
Existing implementation:
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.empty (priority 10)
Hadoop conf-pseudo RPM
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30)
Other Components (like hbase):
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.dist (priority 30)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.dist (priority 30)
We should have consistent configuration for all components
Proposal 1: pseudo configuration as default like hbase/hive
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30) (all pseudo configuration files can be moved to onf.pseudo)
Hadoop conf-pseudo RPM (can be rename to Hadoop Conf-example RPM) like:
/etc/hadoop/conf --> /etc/hadoop/conf.example (priority 40)
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.pseudo (priority 30)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.pseudo (priority 30)
Proposal 2: empty configuration as default like hadoop and pseudo config RPM for each component
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.empty (priority 10)
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.empty (priority 10)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.empty (priority 10)
pseudo configuration RPMs are provided for hadoop/hbase/hive, etc like
Hadoop pseudo RPM:
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30)
Hive pseudo RPM:
/etc/hive/conf --> /etc/hive/conf.pseudo (priority 30)
Hbase pseudo RPM:
/etc/hbase/conf --> /etc/hbase/conf.pseudo (priority 30)
what do you think?
was (Author: rguo):
Hi, Mark,
I agree with you that /etc/<component>/conf is a symlink and can be changed via update-alternative in %post. Maybe I didn't explain very well and pls ignore the above comments.
Existing implementation:
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.empty (priority 10)
Hadoop conf-pseudo RPM
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30)
Other Components (like hbase):
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.dist (priority 30)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.dist (priority 30)
We should have consistent configuration for all components
Proposal 1: pseudo configuration as default like hbase/hive
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30) (all pseudo configuration files can be moved to onf.pseudo)
Hadoop conf-pseudo RPM (can be rename to Hadoop Conf-example RPM) like:
/etc/hadoop/conf --> /etc/hadoop/conf.example (priority 40)
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.pseudo (priority 30)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.pseudo (priority 30)
Proposal 2: empty configuration as default like hadoop
Hadoop Core RPM
/etc/hadoop/conf --> /etc/hadoop/conf.empty (priority 10)
Hbase RPM:
/etc/hbase/conf --> /etc/hbase/conf.empty (priority 10)
Hive RPM:
/etc/hive/conf --> /etc/hive/conf.empty (priority 10)
pseudo configuration RPMs are provided for hadoop/hbase/hive, etc like
Hadoop pseudo RPM:
/etc/hadoop/conf --> /etc/hadoop/conf.pseudo (priority 30)
Hive pseudo RPM:
/etc/hive/conf --> /etc/hive/conf.pseudo (priority 30)
Hbase pseudo RPM:
/etc/hbase/conf --> /etc/hbase/conf.pseudo (priority 30)
what do you think?
> hadoop pseudo configuration can be default configuration
> --------------------------------------------------------
>
> Key: BIGTOP-1339
> URL: https://issues.apache.org/jira/browse/BIGTOP-1339
> Project: Bigtop
> Issue Type: Improvement
> Components: RPM
> Reporter: Guo Ruijing
>
> in existing implementation, hadoop-conf-pseudo-2.2.0.x86_64.rpm is to provide hadoop pseudo configuration. we may take hadoop pseudo configuration as hadoop default configuration and move hadoop pseudo configuration to core/hdfs/yarn rpm.
--
This message was sent by Atlassian JIRA
(v6.2#6252)