Persist Generated File Infile..Row_Generator

[prabu]

Hi DS folks,
please let me know if there is a way available to persist the input file generated by Row generator.i believe it uses a pipe stream to keep the generated records!

2) i can't find the bad file + param file generated, i see only control + log file even when there is a failure[is it the default behavior??]

please let me know

regards,
Prabu

[ray.wurlod]

There are many ways. Possibly the easiest is to follow the Row Generator stage with a Copy stage whose second output sends rows to a file.

[prabu]

There are many ways. Possibly the easiest is to follow the Row Generator stage with a Copy stage whose second output sends rows to a file. ...

Good idea, Ray! BTW, could someone pls explain about the bad file and param not being accessible after a sqlldr failure. Meaning they are not being created, i believe.

pls let me know

regards,
Prabu

[kumar_s]

Pls dont hijack the thread, Create a new post for this purpose.

[kris007]

Isn't this a new post

[kumar_s]

Alright, I was carried away with "sqlldr failure".

[prabu]

Hi DS folks,

2) i can't find the bad file + param file generated, i see only control + log file even when there is a failure[is it the default behavior??]

please let me know

regards,
Prabu

any pointers??

[DSguru2B]

Are you using the DataStage SQL Load stage to BCP the records into the database?
Sounds like an obvious question but i just wanted to be sure.
If you are then there must be an option in the load stage whether to remove the temp files or not. I know for sure that the UDB load stage has that option, i am guessing the SQL load stage should have that option too.

[prabu]

Are you using the DataStage SQL Load stage to BCP the records into the database?
Sounds like an obvious question but i just wanted to be sure.
If you are then there must be an option in the load stage whether to remove the temp files or not. I know for sure that the UDB load stage has that option, i am guessing the SQL load stage should have that option too.

Hi DSguru,

i am using Oracle Enterprise stage. the issue is i cannot see the BAD files generated.

the log file is having the entry as

Code: Select all

1 Row not loaded because all fields were null.

but usually when you invoke the sqlldr with the BAD option, the offending records [that couldn't be loaded] will be moved into the bad records will be moved into the bad records.

but when invoked it thr' DS with the load option , i coulnt see such a BAD file.

regards,
Prabu

[ray.wurlod]

Perhaps an all-NULL row does not get written into a bad file. Check with your Oracle DBA or Oracle documentation. Or experiment outside of DataStage.

[prabu]

Perhaps an all-NULL row does not get written into a bad file. Check with your Oracle DBA or Oracle documentation. Or experiment outside of DataStage. ...

Ray, you are right! all null reocrds go to
discard file

but datastage takes into account only

1)control file
2)log file
3)parfile and
4)bad file

how to pass the discard file also as a parameter. i have tried APT_ORACLE_LOAD_OPTIONS . but it seems to expect a control file. the job fails with

Code: Select all

Expecting keyword LOAD, found "discard".

any thougths??

regards,
Prabu

[prabu]

[ray.wurlod]

I don't currently have access to PX but would imagine that discard file is an optional property of the load stage. If you can find it (in Available Properties to Add), you can use a job parameter to supply its value.