[jira] [Commented] (DRILL-1708) Drillbit hangs when initial connection to zookeeper fails

["Hanifi Gunes (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/DRILL-1708?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14211510#comment-14211510 ] 

Hanifi Gunes commented on DRILL-1708:
-------------------------------------

I have seen here curator has retry policy. We should handle the case where policy does no longer apply. In this case I would think that we should throw an exception and fail.

> Drillbit hangs when initial connection to zookeeper fails
> ---------------------------------------------------------
>
>                 Key: DRILL-1708
>                 URL: https://issues.apache.org/jira/browse/DRILL-1708
>             Project: Apache Drill
>          Issue Type: Bug
>            Reporter: Hanifi Gunes
>            Assignee: Hanifi Gunes
>
> Existing zookeeper cluster coordinator implementation blocks until initial connection succeeds. However, drillbit hangs forever in case connection does not succeed for some reason as there is no retry logic implemented. This issue proposes to implement a cleaner, well defined outcome in case zk connection fails.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)