Post reply

Migrating from SQL server 2000 to 2005

  • August 30, 2009 at 12:13 pm

    #63630

    Hi all,

    Need help in upgradation from SQL Server 2000 to 2005.

    I have migrated the DTS packages to SSIS.

    My package loads from flat files (data files) to my database.

    I am using BULK-INSERT statement.

    I am using non-html format files for blk insert

    My inp data contains spl characters

    Looks something like this

    12,180.oo

    1,290.0

    My destination table has been defined as "float"

    I am facing the folowing error:Bulk load data conversion error (type mismatch or invalid character for the specified codepage)

    Cannot change the datatype to varachar.

    There is no problem while loading in 2000 with the same data and destination as float. But facing the error in 2005:hehe:

    Please help..

    Thank you

    archana

  • August 31, 2009 at 1:42 am

    #1046743

    Hi

    Try inserting a conversion step in you package. Doesnt sound like you are using float data as the source?

    With flat Files, you can also specify the conversion in the import (destination page) step.

    ....

    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
    This thing is addressing problems that dont exist. Its solution-ism at its worst. We are dumbing down machines that are inherently superior. - Gilfoyle

  • August 31, 2009 at 3:34 am

    #1046759

    hi

    Thanks for the reply...

    But can you suggest any changes which can be done without adding conversion stage.

    Can't change the design...:(

    Any idea how this can be done????

    Thank you

    Archana

  • August 31, 2009 at 4:07 am

    #1046771

    Hi, the flat file source you are using, is it a normal text file,or perhaps csv?

    Rerun your job and look for a column name that is giving the issue.

    I know you will have issues dumping txt to float as well. i.e. string type characters.

    I had a similar issue once, but resolved it by changing the source data, and the output column format from float to nvarchar.

    Awaiting your reply.

    -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
    This thing is addressing problems that dont exist. Its solution-ism at its worst. We are dumbing down machines that are inherently superior. - Gilfoyle

  • August 31, 2009 at 5:54 am

    #1046806

    hi..

    Mine is a tab delimted file.

    Yep We can do by changing the source data or by changing the column ' s datatype.

    I have tried this but i dont have the liberty changing either of them.

    Any other option would be helpful

    Thank you

    Archana

Viewing 5 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic. Login to reply