Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Fri, 01 May 2015 @ 14:40:29 GMT


     
  <Prev Next>   <<First <Prev
Next>
Last>>
 


Subj:   Re: Duplicate entries in DBQL
 
From:   Sravan.Bodla

Hi Dave,

No, its not a PREPARE and EXECUTE statement.

Yes, last time I had that doubt and clarified, duplicates are expected for prepare and execute statement like stored procedure etc.

But, this time, we see the duplicates for SELECT and DML statements as well.

Query id duplicate.

Querytext and few other columns will be the same.

But, one of the row will have the ABORT FLAG as 'T'. (Agree that aborted row as ampcputime= 0)


Here is the analysis description from the incident:

"The problem is may happen for multiple client apps with various connectivities(CLI,ODBC,JDBC). The root problem is initializing newly obtained request context with stale data. This can happen due to a timing window where the TSR msg corresponding to a new request comes in while the second phase of PMPC abort is currently in progress for this session. DBQL logging for this type of request is resulting in the duplicate DBQL row being logged to the DBQL table. Please note here that it is not just the QueryID but also the QueryText and many other fields of this row are logged as duplicate of the previous successful request run from this session"


Question is, There can be plenty of sites who are running their system on 14.10 and how can this be a known issue or no workaround. And why is this not fixed or given a temporary solution?

If someone is already facing this issue, How they are handling this or deleting the duplicate entries?


Thanks,

Sravan



     
  <Prev Next>   <<First <Prev
Next>
Last>>
 
 
 
 
 
 
 
 
 
  
  Top Home Privacy Feedback  
 
 
Copyright for the TeradataForum (TDATA-L), Manta BlueSky    
Copyright 2016 - All Rights Reserved    
Last Modified: 23 Jun 2019