tajo-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alvin Henrick (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (TAJO-704) TajoMaster HA
Date Tue, 01 Apr 2014 20:32:20 GMT

     [ https://issues.apache.org/jira/browse/TAJO-704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Alvin Henrick updated TAJO-704:

    Attachment: TAJO-704.Henrick.01.patch.txt

Was going through the list of issues found this interesting one so thought of sharing my views

How about using Akka library actor based model for HA .It solves lot of problem like concurrency
, load balancing and fault tolerance.

I have attached a patch to demonstrate the feature.You can try to kill the TajoMaster and
Supervisor actor will  detect master is down via TajoMasterDownException and will try to restart
the server. 

It is just a sample we can do lot more and better with all your suggestions.

Warm Regards,

> TajoMaster HA
> -------------
>                 Key: TAJO-704
>                 URL: https://issues.apache.org/jira/browse/TAJO-704
>             Project: Tajo
>          Issue Type: Improvement
>          Components: tajo master
>            Reporter: Jaehwa Jung
>            Assignee: Jaehwa Jung
>         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

View raw message