DSXchange

Hello,

I have 4 set of datastage job, which is running in sequence to achieve the desire target valus....
(1) DB2 -->xfm--> Datasets,
(2) Datasets ---->join--->Datasets,
(3) Datasets ---> join-->xfm---> Sequential file
and then
(4) db2 loader to load to target db2 database.

When I check the Sequential file stage, the last one before running db2 loader, I have some decimal values (8,0) which is representing like "00000000."). This creating problem while loading to target database, I am getting error message "cannot be converted to
a PACKED DECIMAL value, but the target column is not nullable" and records are being dropped.

I am trying to convert "00000000." to "0" means supressing zeros so that I can load it to the target database.

Can anyone guide me how to achieve it?

Thanks,

Gaurav

Probably whats happening is that you are trying to load the Null values in a Sequential file. This will not be handled, hence DS is converting the nulls to a default.
I would recommend a peek stage to be added to all the jobs and observe the data there and post your observation.
Thanks.

Zero is not valid to be represented as a packed decimal. Search the forum for "PACKED" "zero" (all terms) to find an environment variable that allows this behaviour to be overridden.

Thanks Ray

Still I am not able find the solution. Wondering if I need edit properties for the particular column (decimal one)..

Regards,
Gaurav

Gaurav.Dave wrote:Thanks Ray

Still I am not able find the solution. Wondering if I need edit properties for the perticular column (decimal one)..

Regards,
Gaurav

you need to do two things .

1> remove the "." from the string when writing it to the file.
2> by default "all zeroes" is not considered as a packed decimal value {atleast in datastage}. So you need to change some property in ur db so that it will accept it.

USE Abs OR DIVIDE THE VALUE BY 1 IN TRANSFORMER