[jira] [Commented] (TAJO-704) TajoMaster HA

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/TAJO-704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14062310#comment-14062310 ] 

ASF GitHub Bot commented on TAJO-704:
-------------------------------------

GitHub user blrunner opened a pull request:

    https://github.com/apache/tajo/pull/77

    TAJO-704: TajoMaster HA

    I implemented TajoMaster HA Manager utilizing HDFS Cluster. I defined a interface named HAManager, and I implemented HAManagerWithHDFS with the interface. 
    
    To use TajoMaster HA, you have to set tajo.master.ha.enable=true at tajo-site.xml. And then, you just run standby master. All master informations will record to hdfs cluster, and all tajo service will refer master service address using hdfs cluster.
    
    For reference, this patch includes a few issues as follows:
    
    - It just implemented one unit test case because of TAJO-942. 
    - If you kill current master and you already ran tsql, you have to restart tsql. because new master doesn't know existing session ids. We need to handle this issue at another ticket.


You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/blrunner/tajo TAJO-704

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/tajo/pull/77.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #77
    
----
commit 47d6c6a32010621b94b31a9871d3a5cd44968667
Author: blrunner <jh...@gruter.com>
Date:   2014-07-15T14:32:32Z

    initial commit: TajoMaster HA Manager utilizing HDFS Cluster.

commit 0b6276738ad4be278d9ab7d222fef78419d42b62
Author: blrunner <jh...@gruter.com>
Date:   2014-07-15T16:25:34Z

    Adding HA elector to client side.

----


> TajoMaster HA
> -------------
>
>                 Key: TAJO-704
>                 URL: https://issues.apache.org/jira/browse/TAJO-704
>             Project: Tajo
>          Issue Type: New Feature
>          Components: tajo master
>    Affects Versions: 0.9.0
>            Reporter: Jaehwa Jung
>            Assignee: Jaehwa Jung
>             Fix For: 0.9.0
>
>         Attachments: TAJO-704.Henrick.01.patch.txt, TajoMasterHAdraft.pdf
>
>
> TajoMaster is a Single Point of Failure in a Tajo Cluster because TajoMaster is the central controlling entity for all components of the Tajo system. TajoMaster failure prevents clients from submitting new queries to the cluster, and results in the disruption of the ability to run insert overwrite queries because the TajoWorker can’t apply its statistical information to CatalogStore. Therefore, the high-availability (HA) of TajoMaster is essential for the high-availability of Tajo generally. 



--
This message was sent by Atlassian JIRA
(v6.2#6252)