hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jing Zhao (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-10535) Make the retry numbers in ActiveStandbyElector configurable
Date Wed, 23 Apr 2014 20:48:16 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-10535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jing Zhao updated HADOOP-10535:
-------------------------------

    Description: Currently in ActiveStandbyElector, when its zookeeper client cannot successfully
communicate with ZooKeeper service, the retry number is hard coded to 3 (ActiveStandbyElector#NUM_RETRIES).
After retrying 3 times a fatal error will be thrown and the ZKFC will quit. It will be better
to make the retry times configurable.  (was: Currently in ActiveStandbyElector, when its zookeeper
client cannot successfully communicate with ZooKeeper server, the retry number is hard coded
to 3 (ActiveStandbyElector#NUM_RETRIES). After retrying 3 times a fatal error will be thrown
and the ZKFC will quit. It will be better to make the retry times configurable.)

> Make the retry numbers in ActiveStandbyElector configurable
> -----------------------------------------------------------
>
>                 Key: HADOOP-10535
>                 URL: https://issues.apache.org/jira/browse/HADOOP-10535
>             Project: Hadoop Common
>          Issue Type: Improvement
>            Reporter: Jing Zhao
>            Assignee: Jing Zhao
>            Priority: Minor
>
> Currently in ActiveStandbyElector, when its zookeeper client cannot successfully communicate
with ZooKeeper service, the retry number is hard coded to 3 (ActiveStandbyElector#NUM_RETRIES).
After retrying 3 times a fatal error will be thrown and the ZKFC will quit. It will be better
to make the retry times configurable.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message