[
https://issues.apache.org/jira/browse/HDDS-2701?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17759828#comment-17759828
]
Sumit Agrawal commented on HDDS-2701:
-------------------------------------
Currently, we are supporting FILE_PER_BLOCK default strategy, so above may not
be applicable. Additionally, *HDDS-8324* has handled making cache to be present
till written to one of follower.
Further improvement needs to be handled HDDS-9222 for caching.
> Avoid read from temporary chunk file in datanode
> ------------------------------------------------
>
> Key: HDDS-2701
> URL: https://issues.apache.org/jira/browse/HDDS-2701
> Project: Apache Ozone
> Issue Type: Bug
> Components: Ozone Datanode
> Reporter: Lokesh Jain
> Assignee: Lokesh Jain
> Priority: Major
> Labels: TriagePending
>
> Currently we try reading chunk data from the temp file if chunk file does not
> exist. The fix was added in HDDS-2372 due to race condition between
> readStateMachineData and writeStateMachineData in ContainerStateMachine.
> After HDDS-2542 is fixed the read from the temp file can be avoided by making
> sure that chunk data remains in cache until the chunk file is generated.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]