[ 
https://issues.apache.org/jira/browse/IOTDB-3720?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17563848#comment-17563848
 ] 

任宇华 commented on IOTDB-3720:
----------------------------

[https://apache-iotdb.feishu.cn/docx/doxcnotnSFMcrwMhQd23ZvPp88e]

> Unify retry logic of AsyncClientPool in ConfigNode
> --------------------------------------------------
>
>                 Key: IOTDB-3720
>                 URL: https://issues.apache.org/jira/browse/IOTDB-3720
>             Project: Apache IoTDB
>          Issue Type: Improvement
>            Reporter: Yongzao Dan
>            Assignee: 任宇华
>            Priority: Minor
>             Fix For: 0.14.0
>
>
> Currently, none of ConfigNode's AsyncClientPool interfaces have retry logic 
> except for createRegions. The createRegions interface uses CountDownLatch and 
> BitSet to construct the retry logic. Although this algorithm can fulfill the 
> requirements, it is difficult to reuse and causes the code inside the 
> interface to be too long. We need a uniform way to elegantly retry RPC 
> requests in AsyncClientPool.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to