Problem with APT_STRING_PADCHAR

[Nageshsunkoji]

Hi Folks,

I am facing a strange problem with $APT_STRING_PADCHAR. I am using this varible for my DB2 EE Load job. It's working fine for my Load job. But, the problem is , when i Tried to set this variable in the sequence. This variable is affecting the other jobs(which don't have $APT_STRING_PADCHAR variable) which are present in the sequence. It's not filtering the values in the filter stage for CHAR columns and the strange thing is filter is happening with integer columns.

Please suggest me to sort out from this problem.

Is the varible $APT_STRING_PADCHAR setting in the sequence will affect the other jobs, even though those jobs are not using this environment variable.

Thanks & Regards,
Nagesh.

[pavankvk]

$APT_STRING_PADCHAR is a global level variable applicable for the whole project and all the jobs in it.

[ArndW]

...even though those jobs are not using this environment variable...

The fact that the environment variable is not explicitly declared does not mean that the EE jobs don't use it. The same applies to ALL environment variables. If they are not explicitly set in the parameters of a job then their values are inherited.

[kumar_s]

What is the value assigned to the variable? What is the filter condition and what is getting escaped?

[Nageshsunkoji]

Arnd,

Thanks for your reply.

My problem is the APT_STRING_PADCHAR is affecting the jobs, which don't have DB2 EE Stages.

My Job flow :

first Job

DB2 API STAGE ..................Dataset

Second Job



DATASET ( this one from First Job)..............FILTER .................. DATASET

So, the above two jobs are part of my sequence.These two jobs are not using any environment variable.

But, if I used the APT_STRING_PADCHAR in the sequence, It is affecting the above two jobs also.
Records are not filtering the CHAR values.

Regards
Nagesh.

[kumar_s]

IF you tend to change the default value of the variable after assigning in the sequence, yes it will affect.

[ArndW]

The APT_STRING_PADCHAR is used anytime a string needs to get padded and not only for DB/2 EE stages. So if you changed the pad char to "x" and do a transform or implicit conversion from a varchar(3) to a CHAR(10) column then it will use 'x'. If this is wrong in the specific job, add APT_STRING_PADCHAR to your job's parameters and give it the value you want.

[Nageshsunkoji]

The problem with regards to $APT_STRING_PADCHAR is coming in specific project. In development project the sequence using this parameter works fine. However, in the sys test project the sequence is giving problem. Is there any other enviornment variable which is dependent on $APT_STRING_PADCHAR or is it something to do with data problem? Help on this would be very helpful.

[ArndW]

Nagesh, there really isn't a (DataStage) problem here.

When DS needs to perform padding it will use the APT_STRING_PADCHAR value unless you override it. The job will use the current APT_STRING_PADCHAR, that value is set to a default value at project level but can be changed or set at a job level.

It would seem that your development and production default settings are different. There are no other environment variables that are dependant on this one. There are other padding settings out there (I recall having had issues with the ORACLE one in the past). But this one is what controls the default padding of strings in PX jobs.

[kumar_s]

And more over, this variable is not required to be inserted into the any job or sequence, unless it is required to be changed. Is there any specific reason that, the use of variable in the sequence?