Re: weirds problems after upgrade from 7.5.00.18 to 7.5.00.19

Jean-Michel OLTRA wrote:
> * on PROD instance, the first insert is performed and the session commited
>   after the second one.
>   There is no error but the second phase seems to have been skipped
>   (no inserts). Remember, I work with the same scripts and dbproc.
> # on both instances, ireport cannot find any row by pk on the first table
>   wrote by script one, but can find rows with another column in the query.
>
>   [22:38]jm@espinasse > ireport -d TEST
>   SAP DB Terminal Report 0.3.3
>   Type 'help' for help.
>   An empty line repeats the last command
>   ==> select * FROM FLUXFINANCIER where NOPIECE=696
>   
> |NOPIECE|NOFACTURE|DATEFACTURE|IDETABLISSEMENT|DATEREGLEMENT|IDMODEREGLEMENT|NOCOMPTE|
>   
> +-------+---------+-----------+---------------+-------------+---------------+--------+
>   <<<End of Cursor>>>
>   ==> select * FROM FLUXFINANCIER where NOFACTURE='retro-20041122'
>   
> |NOPIECE|NOFACTURE|DATEFACTURE|IDETABLISSEMENT|DATEREGLEMENT|IDMODEREGLEMENT|NOCOMPTE|
>   
> +-------+---------+-----------+---------------+-------------+---------------+--------+
>   |696|retro-20041122|20041122|3409-001|?|?|?|
>   ==> Q

There is a bug with this kernel version where some some SELECTs don't 
seem to find anything. This seems to occur only with the Python driver 
as rows from cursors are read using FETCH RELATIVE and not FETCH NEXT as 
most of the other interfaces do. It seems to occur only when rows a 
qualified by a key column. This will be fixed in 7.5.00.21. You can work 
around this problem by issuing a resultset.first () before the first 
.next ().

Daniel Dittmar

--
Daniel Dittmar
SAP Labs Berlin
daniel.dittmar@xxxxxxx


--
MaxDB Discussion Mailing List
For list archives: http://lists.mysql.com/maxdb
To unsubscribe:    http://lists.mysql.com/maxdb?unsub=gcdm-maxdb@xxxxxxxxxxx