[ 
https://issues.apache.org/jira/browse/HDDS-7537?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Krishna Kumar Asawa reassigned HDDS-7537:
-----------------------------------------

    Assignee: Nandakumar

> SCM should close pipelines upon restart if pipeline limit is reached
> --------------------------------------------------------------------
>
>                 Key: HDDS-7537
>                 URL: https://issues.apache.org/jira/browse/HDDS-7537
>             Project: Apache Ozone
>          Issue Type: Bug
>            Reporter: Wei-Chiu Chuang
>            Assignee: Nandakumar
>            Priority: Major
>
> Let's say ozone.scm.datanode.pipeline.limit is initially 10, and so each 
> DataNode is on 10 open pipelines.
> If we reduce the limit to 2, restart the cluster, because pipeline info is 
> persisted in the SCM db, even though the number of open pipelines per 
> DataNode is 10 which exceeds the new limit of 2, it does not proactively 
> close the pipelines and bring the number back to 2.
> Because the number of open pipelines are too many, the DN reserves too much 
> committed space for open containers, Ozone can no longer write to that DN 
> because the available space excluding committed space is insufficient.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to