DSXchange

My Source have sql server datatype has number,target have sql server datatype has sql server.When i am loading soure to tartget,data will be truncated.
Ex:Source has 16.8 target loaded to 16.799992372605.
i am not use any convertion function.Please let me know have idea in this.

kethya11 wrote:target have sql server datatype has sql server.

Anyway, perhaps the replies here can help, even if they are about Oracle.

Use the Sever job to load and compare the value.

kethya11 wrote:My Source have sql server datatype has number,target have sql server datatype has sql server.When i am loading soure to tartget,data will be truncated.
Ex:Source has 16.8 target loaded to 16.799992372605.
i am not use any convertion function.Please let me know have idea in this.

Can you please report the SQL data type, the precision and the scale for this column on both source and target?

"sql server" is not a data type.

Please advise also whether you imported the table definitions from source and from target before constructing this job?

Finally, please advice with which stage types you are connecting to the SQL Server database(s).

Hi Ray,

Thank for your reply

i am wrongly posted datatype ,datatype has float .Its no precision and scale values.
After construct the job am imported the meta data.I a, I am using the odbc connecter stage.

Please help me.

The float and dfloat data types are not guaranteed to store data to the final significant digits accurately. Research the IEEE specifications for storing these data types for the reasons why. I'd suggest you specify at least a scale value (ideally a precision value also), and perhaps treat the data type as decimal (though you may have to ignore/demote metadata mismatch warnings).

To load the float value i use Server job and not PX. PX always give different value whereas server is not.

Thanks for giving reply to all

use Double data type.

Double is the same as dfloat.