[ 
https://issues.apache.org/jira/browse/HDDS-3548?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

runzhiwang updated HDDS-3548:
-----------------------------
    Description: 
Multi-Raft: 3 pipelines each datanode, the latest commit is 
https://github.com/apache/hadoop-ozone/commit/099e80858f4382e2a6f9887f548a19e1f6de8ddd

The reason is datanode can not connect to scm, then PipelineReports send failed.

*{color:#DE350B}1.  LinkedList @ 0x77053df98 occupy 1.8GB memory{color}*

!image-2020-05-06-17-11-43-422.png!

*{color:#DE350B} 2. The item in LinkedList @ 0x77053df98 are most 
PielineReportsProto held by StateContext::reports.{color}*

!image-2020-05-06-17-16-02-519.png!

*{color:#DE350B}3. 3107 instances of PielineReportsProto occupy 1.8GB 
memory{color}*

!image-2020-05-06-17-06-21-189.png!


  was:
Multi-Raft: 3 pipelines each datanode, the latest commit is 
https://github.com/apache/hadoop-ozone/commit/099e80858f4382e2a6f9887f548a19e1f6de8ddd

The reason is datanode can not connect to scm, then 

*{color:#DE350B}1.  LinkedList @ 0x77053df98 occupy 1.8GB memory{color}*

!image-2020-05-06-17-11-43-422.png!

*{color:#DE350B} 2. The item in LinkedList @ 0x77053df98 are most 
PielineReportsProto held by StateContext::reports.{color}*

!image-2020-05-06-17-16-02-519.png!

*{color:#DE350B}3. 3107 instances of PielineReportsProto occupy 1.8GB 
memory{color}*

!image-2020-05-06-17-06-21-189.png!



> Datanode memory leak because too many PipelineReportsProto
> ----------------------------------------------------------
>
>                 Key: HDDS-3548
>                 URL: https://issues.apache.org/jira/browse/HDDS-3548
>             Project: Hadoop Distributed Data Store
>          Issue Type: Sub-task
>            Reporter: runzhiwang
>            Assignee: runzhiwang
>            Priority: Major
>         Attachments: image-2020-05-06-17-06-21-189.png, 
> image-2020-05-06-17-11-43-422.png, image-2020-05-06-17-16-02-519.png
>
>
> Multi-Raft: 3 pipelines each datanode, the latest commit is 
> https://github.com/apache/hadoop-ozone/commit/099e80858f4382e2a6f9887f548a19e1f6de8ddd
> The reason is datanode can not connect to scm, then PipelineReports send 
> failed.
> *{color:#DE350B}1.  LinkedList @ 0x77053df98 occupy 1.8GB memory{color}*
> !image-2020-05-06-17-11-43-422.png!
> *{color:#DE350B} 2. The item in LinkedList @ 0x77053df98 are most 
> PielineReportsProto held by StateContext::reports.{color}*
> !image-2020-05-06-17-16-02-519.png!
> *{color:#DE350B}3. 3107 instances of PielineReportsProto occupy 1.8GB 
> memory{color}*
> !image-2020-05-06-17-06-21-189.png!



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: ozone-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: ozone-issues-h...@hadoop.apache.org

Reply via email to