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

Posted to notifications@iotdb.apache.org by "QiangShaowei (Jira)" <ji...@apache.org> on 2023/09/21 09:30:00 UTC

[jira] [Assigned] (IOTDB-6168) ConfigNode register retry logic does not worked

     [ https://issues.apache.org/jira/browse/IOTDB-6168?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

QiangShaowei reassigned IOTDB-6168:
-----------------------------------

    Assignee: QiangShaowei

> ConfigNode register retry logic does not worked 
> ------------------------------------------------
>
>                 Key: IOTDB-6168
>                 URL: https://issues.apache.org/jira/browse/IOTDB-6168
>             Project: Apache IoTDB
>          Issue Type: Bug
>            Reporter: QiangShaowei
>            Assignee: QiangShaowei
>            Priority: Minor
>
> ConfigNode register itself to cluster has retry logic, but does not worked.
> {code:java}
> ConfigNode.java
> for (int retry = 0; retry < STARTUP_RETRY_NUM; retry++) {
>   TSStatus status;
>   TConfigNodeRegisterResp resp = null;
>   Object obj =
>       SyncConfigNodeClientPool.getInstance()
>           .sendSyncRequestToConfigNodeWithRetry(
>               targetConfigNode, req, ConfigNodeRequestType.REGISTER_CONFIG_NODE); 
> ....{code}
> There exist two retry logic: One is in ConfigNode.java for loop, another is in SyncConfigNodeClientPool.getInstance() .sendSyncRequestToConfigNodeWithRetry
> but when the second retry reach max times, the for loop retry logic in ConfigNode does not worked.
>  
> The reason is for loop does not check TSStatusCode.INTERNAL_REQUEST_RETRY_ERROR from SyncConfigNodeClientPool, so throw StartupException abort the for loop



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