[
https://issues.apache.org/jira/browse/HDDS-3133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17723517#comment-17723517
]
Sumit Agrawal commented on HDDS-3133:
-------------------------------------
[~umamahesh] ,
objectId in Ratis: its created using epoch and transactionId
objectId in non-ratis: objectId is created using epoch and transactionId here
is auto-increment value, so this also seems unique
[https://github.com/apache/ozone/blob/bdd3f4e85155070e39135adbbe0be3dfbf5ee138/hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/protocolPB/OzoneManagerProtocolServerSideTranslatorPB.java#L278]
So objectId can be used as fileId.
OzoneFileStatus already have KeyInfo and which in-turn have objectId. So this
can be used for integration on hive.
> Add export objectIds in Ozone as FileIds to allow LLAP to cache the files
> -------------------------------------------------------------------------
>
> Key: HDDS-3133
> URL: https://issues.apache.org/jira/browse/HDDS-3133
> Project: Apache Ozone
> Issue Type: Bug
> Components: Ozone Filesystem, Ozone Manager
> Reporter: Mukul Kumar Singh
> Priority: Critical
>
> Hive LLAP makes use of the fileids to cache the files data. Ozone's objectIds
> need to be exported as fileIds to allow the caching to happen effectively.
> https://github.com/apache/hive/blob/master/ql/src/java/org/apache/hadoop/hive/ql/io/HdfsUtils.java#L65
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
