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

Posted to notifications@iotdb.apache.org by "周沛辰 (Jira)" <ji...@apache.org> on 2023/02/14 08:17:00 UTC

[jira] [Commented] (IOTDB-5527) 0.13.4 Compaction OOM

    [ https://issues.apache.org/jira/browse/IOTDB-5527?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17688359#comment-17688359 ] 

周沛辰 commented on IOTDB-5527:
----------------------------

OOM occurs during system operation, and there are inner space compaction and query at the same time, but no cross space compaction, so it is difficult to accurately locate the OOM problem caused by compaction. In addition, after the system restarts and recovers from OOM, the recover process of  compaction reports an exception. This problem has been fixed in 0.13.4. See IOTDB-5286.

> 0.13.4 Compaction OOM
> ---------------------
>
>                 Key: IOTDB-5527
>                 URL: https://issues.apache.org/jira/browse/IOTDB-5527
>             Project: Apache IoTDB
>          Issue Type: Bug
>            Reporter: 周沛辰
>            Assignee: 周沛辰
>            Priority: Major
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> Find a scenario where the 0.13.4 compaction will lead to memory OOM: the test environment for importing historical data, the JVM memory of about 10G, the write throughput is relatively high, and the out-of-order ratio is not low. After using this data to start in a small memory environment, it will directly OOM.
>  
> Test Data：https://cloud.tsinghua.edu.cn/d/77627e3d9f48450caf0a/



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