[
https://issues.apache.org/jira/browse/HBASE-17272?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15729001#comment-15729001
]
Joep Rottinghuis commented on HBASE-17272:
------------------------------------------
Thanks for clarifying and documenting this!
In conf/hbase-site you still have a value in hbase.rootdir that appears to
point to an actual HDFS url. Did you mean to
"hdfs://namenode.example.org:9000/hbase" as in the description of that
property, or perhaps "hdfs://namenode.example.org:8020/hbase" as described in
/src/main/asciidoc/_chapters/configuration.adoc ?
> Doc how to run Standalone HBase over an HDFS instance; all daemons in one JVM
> but persisting to an HDFS instance
> ----------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-17272
> URL: https://issues.apache.org/jira/browse/HBASE-17272
> Project: HBase
> Issue Type: Task
> Components: documentation
> Reporter: stack
> Attachments: 17272.txt
>
>
> It is possible to run hbase in standalone mode -- all daemons in the one JVM
> (Master+RegionServers+ZooKeeper) -- but persisting to HDFS rather than the
> usual local filesystem. That this works was news to me. This configuration is
> wanted for the default deploy of the timeline server v2. The standalone mode
> hbase shoudl be good enough for light loading. The persistence to HDFS makes
> it so data can live across the coming and goings of nodes.
> This issue is about doc'ing how to setup this config.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
