You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-issues@hadoop.apache.org by "Xuan Gong (JIRA)" <ji...@apache.org> on 2015/10/22 19:36:27 UTC

[jira] [Commented] (HADOOP-12503) ActiveStandbyElector#initialization do not have retry on establishing the zookeeper connection.

    [ https://issues.apache.org/jira/browse/HADOOP-12503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14969514#comment-14969514 ] 

Xuan Gong commented on HADOOP-12503:
------------------------------------

The solution here could be: ActiveStandbyElector could take a additional constructor argument to give the users option on whether they want the retry on establishing zk connection or not. So, we will not break any previous behavior, especially for HDFS ZKFC.

We will fix this together in https://issues.apache.org/jira/browse/YARN-4243

> ActiveStandbyElector#initialization do not have retry on establishing the zookeeper connection.
> -----------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-12503
>                 URL: https://issues.apache.org/jira/browse/HADOOP-12503
>             Project: Hadoop Common
>          Issue Type: Bug
>            Reporter: Xuan Gong
>            Assignee: Xuan Gong
>
> Currently, in ActiveStandbyElector#initialization, we do not have retry on establishing the zookeeper connection. Some of the components which extend ActiveStandbyElector, such as EmbeddedElectorService in YARN ResourceManager, need the retry. 



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