hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Karthik Kambatla (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-10182) ZKFailoverController should implicitly ensure that ZK is formatted
Date Thu, 26 Dec 2013 20:52:51 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-10182?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13857068#comment-13857068
] 

Karthik Kambatla commented on HADOOP-10182:
-------------------------------------------

Formatting without an explicit request seems like a bad idea - a wrong configuration could
easily end up formatting the znode of a different cluster. At the very least, if we decide
to go this route, should have a config to disable/ enable this.

> ZKFailoverController should implicitly ensure that ZK is formatted
> ------------------------------------------------------------------
>
>                 Key: HADOOP-10182
>                 URL: https://issues.apache.org/jira/browse/HADOOP-10182
>             Project: Hadoop Common
>          Issue Type: Bug
>          Components: ha
>    Affects Versions: 2.3.0
>            Reporter: Ivan Mitic
>            Assignee: Ivan Mitic
>
> Currently, a manual ZK format step is necessary before ZKFailoverController can join
leader election. This generally adds overhead on what is needed to configure an HA cluster.
My proposal is to always ensure that ZK is formatted during startup of the ZKFailoverController.
If folks do not want to make this a default, we can put it under a config.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message