hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sankar Hariappan (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HIVE-19435) Incremental replication cause data loss if table with same name is dropped followed by create and insert-into with different partition type.
Date Mon, 07 May 2018 08:34:00 GMT

     [ https://issues.apache.org/jira/browse/HIVE-19435?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sankar Hariappan updated HIVE-19435:
------------------------------------
    Description: 
If the incremental dump have drop of partitioned table followed by create/insert on non-partitioned
table with same name, doesn't replicate the data. Explained below.

Let's say we have a partitioned table T1 which was already replicated to target.

DROP_TABLE(T1)->CREATE_TABLE(T1) (Non-partitioned) -> INSERT(T1)(10) 

After REPL LOAD, T1 doesn't have any data.

Same is valid for no-partitioned to partitioned case as well.

 

  was:
If the incremental dump have drop of partitioned table followed by create/insert on non-partitioned
table with same name, doesn't replicate the data. Explained below.

Let's say we have a partitioned table T1 which was already replicated to target.

DROP_TABLE(T1)->CREATE_TABLE(T1) (Non-partitioned) -> INSERT(T1)(10) 

After REPL LOAD, T1 doesn't have any data.

 


> Incremental replication cause data loss if table with same name is dropped followed by
create and insert-into with different partition type.
> --------------------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-19435
>                 URL: https://issues.apache.org/jira/browse/HIVE-19435
>             Project: Hive
>          Issue Type: Bug
>          Components: HiveServer2, repl
>    Affects Versions: 3.0.0
>            Reporter: Sankar Hariappan
>            Assignee: Sankar Hariappan
>            Priority: Major
>              Labels: DR, replication
>             Fix For: 3.0.0, 3.1.0
>
>
> If the incremental dump have drop of partitioned table followed by create/insert on
non-partitioned table with same name, doesn't replicate the data. Explained below.
> Let's say we have a partitioned table T1 which was already replicated to target.
> DROP_TABLE(T1)->CREATE_TABLE(T1) (Non-partitioned) -> INSERT(T1)(10) 
> After REPL LOAD, T1 doesn't have any data.
> Same is valid for no-partitioned to partitioned case as well.
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message