[
https://issues.apache.org/jira/browse/KYLIN-2006?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15481066#comment-15481066
]
kangkaisen commented on KYLIN-2006:
-----------------------------------
The main solution is changing the cube level lock for the server level lock.
To be specific:
1 When submit or resume a job, the job server need to get the cube level lock
and write its serverName to the znode.
2 Before the job scheduler schedule the job, it need to decide the job server
whether or not have the cube level lock.
3 After the job succeed, fail or discard, the job server will release the cube
level lock.
4 This feature will add a new scheduler class and a new zookeeper lock class.
5 This feature is configurable and default isn't enable.
> Make job build server distributed
> ---------------------------------
>
> Key: KYLIN-2006
> URL: https://issues.apache.org/jira/browse/KYLIN-2006
> Project: Kylin
> Issue Type: New Feature
> Components: Job Engine
> Reporter: kangkaisen
> Assignee: kangkaisen
>
> Currently, the Kylin job build server is single-point。
> In order to make Kylin job build server more extensible, available and
> reliable, we should support distributed job build server.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
