hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eugene Koifman (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-17698) FileSinkDesk.getMergeInputDirName() uses stmtId=0
Date Wed, 04 Oct 2017 21:34:00 GMT

    [ https://issues.apache.org/jira/browse/HIVE-17698?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16192069#comment-16192069
] 

Eugene Koifman commented on HIVE-17698:
---------------------------------------

In fact,
FileSinkOperator.initializeBucketPaths(int filesIdx, String taskId, boolean isNativeTable,
        boolean isSkewedStoredAsSubDirectories)
calls           
String subdirPath = AcidUtils.deltaSubdir(txnId, txnId, stmtId);
which means it should already generate a different delta/ (by suffix) for each FileSinkOperator


> FileSinkDesk.getMergeInputDirName() uses stmtId=0
> -------------------------------------------------
>
>                 Key: HIVE-17698
>                 URL: https://issues.apache.org/jira/browse/HIVE-17698
>             Project: Hive
>          Issue Type: Sub-task
>          Components: Transactions
>            Reporter: Eugene Koifman
>
> this is certainly wrong for multi statement txn but may also affect writes from Union
All queries if these are made to follow full Acid convention
> _return new Path(root, AcidUtils.deltaSubdir(txnId, txnId, 0));_



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message