[
https://issues.apache.org/jira/browse/HDDS-12902?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Attila Doroszlai updated HDDS-12902:
------------------------------------
Fix Version/s: 2.1.0
Resolution: Fixed
Status: Resolved (was: Patch Available)
> CloseContainerCommandHandler and ECReconstructionCoordinator should shut down
> their thread executor gracefully
> --------------------------------------------------------------------------------------------------------------
>
> Key: HDDS-12902
> URL: https://issues.apache.org/jira/browse/HDDS-12902
> Project: Apache Ozone
> Issue Type: Bug
> Reporter: Wei-Chiu Chuang
> Assignee: Peter Lee
> Priority: Minor
> Labels: pull-request-available
> Fix For: 2.1.0
>
>
> ECReconstructionCoordinator has ecReconstructReadExecutor which creates
> ThreadPoolExecutor as user-thread, not daemon thread.
> It doesn't cause problem today, as this thread is inside DataNode, and
> DataNode shuts down when the process shuts down.
> In general a ThreadPoolExecutor should either:
> (1) be a daemon thread, or
> (2) a user-thread, and shut down explicitly when the parent object closes.
> It looks like this bad pattern happens in a few other places:
> CloseContainerCommandHandler, FinalizationManager,
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
