Teradata multiload 1 record gets dropped

alankyho · Post by **alankyho** » Mon Jun 18, 2007 9:08 am

My source dataset has 60097 records as input to Teradata multiload stage. But what I get for inserts is only 60096 records on the table.
There is no duplicates kick out on UV report and no rows on ET report.

I'm using keys on AR_ID, DW_SRVC_ID and CAPTR_DT and there are no duplicates on combination of these keys. The table has no primary key but only primary index (AR_ID).

Can anyone know of any reasons why 1 record gets dropped from doing inserts to the table?

samsuf2002 · Post by **samsuf2002** » Mon Jun 18, 2007 10:28 am

Check out it might be the header....

chulett · Post by **chulett** » Mon Jun 18, 2007 10:33 am

As noted, make sure you haven't accidentally checked the 'First line is column names' option which basically pitches the first record in the file.

ray.wurlod · Post by **ray.wurlod** » Mon Jun 18, 2007 3:18 pm

Unlikely to be "first line is column names" from a Data Set.

What is your job design? Please use "ASCII art" within Code tags - it makes it so much easier to read.

What row counts are reported on each link?

Enable APT_RECORD_COUNTS (perhaps as a job parameter) so that the player processes log their row counts.

Is there some reason you're preferring the MultiLoad stage to the Teradata Enterprise stage? Are you specifying a Report file on the MultiLoad/TPump tab?

chulett · Post by **chulett** » Mon Jun 18, 2007 5:34 pm

Sorry, guess I read 'dataset' in way too generic of a manner. Doh.