DSXchange

You're not forced to update it but you do need to use it, typically in the 'wher'e clause if nothing else.

If you don't need a column in the DML, don't send it to the stage.

I would try stripping all of the non-numeric values and then check for zero:
(assuming no other transformations are needed)

First, it is "orchadmin" and secondly you've have to give us some idea what "not working" means before anyone could help you without just guessing what might be wrong.

Please post the other error messages logged along with it, they generally don't happen all on their lonesome.

Why not simply search for it? Use the search functionality if you are on Windows or the 'find' command from UNIX, it shouldn't be all that hard to turn it up.

Are you sure that is the return status? Or does it simply echo that to the screen while the job is running? I'd also be curious exactly how you are capturing the status into your $Run_Stat_Val variable - can you post that segment of the script?

Odd... don't recall "wide character" ever being something posted here before. Not sure if this helps at all but there's a stub wiki article on the subject.

Well... you might be surprised how many people would rather come here and ask than read any of the documentation or try to work something out on their own, so just wanted to get that out of the way since it seemed from your post like you might not have been thru the docs yet. And it's not like I use...

Explain your job design - basically list the stages used and what role each is playing in the job.

It would behoove you greatly to read the documentation that ships with the product, specifically the Parallel Job Developer's Guide pdf. There is a chapter on the Lookup Stage there as well as the related Join Stage and Merge Stage. You need to understand how they work in order to be successful desi...

Please start your own post rather than 'hijack' someone else's.

Before 8.x users are just normal O/S users so create them there. Roles are handled via the Administrator using the 'Permissions' tab.

The ODBC stage is the only stage that allows that, AFAIK.

If the stage generates the SQL then yes, it will already use TO_CHAR. And you need it, one way or the other, to make this work properly. Lastly you can't use TO_TIMESTAMP as you're not dealing with an Oracle TIMESTAMP column, but rather a DATE - which just so happens to carry a time portion in Orac...

That just looks like a Pivot to me, why not perform the pivot of using the maximum number of output columns and then trim null results post pivot?