DSXchange

There is no criteria for which duplicate is choosen - it is just the first one encounted that the match will be performed on.

So, unless you want multiple rows in the output, then you should remove duplicates in either a sort or remove duplicate stage, where you can define the criteria

multiply with -1

If you look into your Oracle DB - then how does your source data appear? (if you have a lot of ??? there I guess you know why you have it in your target also) Speciel characters are many times not viewable using the "view" function in datastage. Try to output the result to a flat file and ...

The only way to read a state file is using the SKG stage and dumping the output to the log.

What is the source stage for the reference data (ID column)? Seq file stage? Connector? Are you able to view data from there?

Have you tried to generate a new state file for the job (one that has not been updated by the SCD stage) to check if that removes the warning?

UltraEdit (hex editor) is available as a free triel version here http://www.ultraedit.com/downloads.html

I can't answer why - but I know it is not supported in a read operation from a sequential file either. But it must be defined as a smallint, int or NChar(1) in order to get the right result.

If you want a null value from your lookup operation the field needs to be nullable on both the input and output side - otherwise it will default to 0 for a integer

you can use the Padstring function on your string. It enables you to define a optional pad character and lenght

For me it looks like you are trying to put a varchar into a datetime... where exactly are you doing your type conversion? - in a transformer?

The condition should be alright.

Try to set the lookup failure options to continue and view the result in a peak stage - Do you have a lookup match where your condition is true?

Yes... you can harness the power of DB sorting in your DS job... HOWEVER... you'll need to consider... #1. If you have to join to sources then you need to have your data key partitioned, and a re-partitioning can destroy your sort order. #2. Database collation vs. DS project collation. The sorting o...

You need to tell DS what your timestamp from the source looks like...

The format for '8-29-2011 5:26:03AM' is not '%yyyy-%mm-%dd %hh:%nn:%ss'