[ 
https://issues.apache.org/jira/browse/TRAFODION-2722?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16133708#comment-16133708
 ] 

ASF GitHub Bot commented on TRAFODION-2722:
-------------------------------------------

Github user selvaganesang commented on a diff in the pull request:

    
https://github.com/apache/incubator-trafodion/pull/1211#discussion_r134066270
  
    --- Diff: core/sql/src/main/java/org/trafodion/sql/HBaseClient.java ---
    @@ -648,6 +648,8 @@ public boolean truncate(String tblName, boolean 
preserveSplits, long transID)
                }
                else {
                   TableName tableName = TableName.valueOf(tblName);
    +              if (admin.isTableEnabled(tableName))
    +                  admin.disableTable(tableName);
    --- End diff --
    
    Why do we disable the table before truncate? And when will it be enabled. 


> change purgedata to use hbase truncate api instead of drop/create
> -----------------------------------------------------------------
>
>                 Key: TRAFODION-2722
>                 URL: https://issues.apache.org/jira/browse/TRAFODION-2722
>             Project: Apache Trafodion
>          Issue Type: Improvement
>            Reporter: Anoop Sharma
>            Assignee: Anoop Sharma
>            Priority: Minor
>
> currently, traf purgedata command runs by first dropping underlying hbase
> objects for a table and its associated indexes, and then recreating them.
> At recreation time, the original hbase options and keys are used.
> drop/create was done during early days of traf when hbase didn't provide an 
> api
> to truncate objects.
> Now that the truncate api is available, this jira proposes changing code to 
> use it
> instead of drop/create. This would improve performance as well as avoid the 
> need
> to recreate using the original options.



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

Reply via email to