[ 
https://issues.apache.org/jira/browse/YARN-4243?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14953332#comment-14953332
 ] 

Xuan Gong commented on YARN-4243:
---------------------------------

[~rohithsharma] Thanks for the review.
bq. And method reEstablishSession() can be reused rather duplicating same logic 
over embedded electors. Instead of overriding createConnection() method, 
reEstablishSession() method can be used in ActiveStandByElector constructor.I'd 
prefer to make change in hadoop-common rather in embedded elector service.

This will affect the HDFS ZKFS, and they do not want the retry on 
initialization.

bq. While initializing Elector service createConnection will retry as per 
configured value i.e maxRetryNum say 10. But if session is closed and 
reestablished then number of retry count will be maxRetryNum * maxRetryNum i.e 
10*10=100 times.

I am not sure if I understand correctly. If we set the maxRetryNum as 10, and 
zk connect itself would do some retries (10times), the total is 10*10.

> Add retry on establishing Zookeeper conenction in 
> EmbeddedElectorService#serviceInit
> ------------------------------------------------------------------------------------
>
>                 Key: YARN-4243
>                 URL: https://issues.apache.org/jira/browse/YARN-4243
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: resourcemanager
>            Reporter: Xuan Gong
>            Assignee: Xuan Gong
>         Attachments: YARN-4243.1.patch
>
>
> Right now, the RM would shut down if the zk connection is down when the RM do 
> the initialization. We need to add retry on this part



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

Reply via email to