DSXchange

Unfortunately, the job isn't really "generic" unless it uses RCP and the need to call a routine for every column rules that out, as noted. I'm guessing about as generic as you could get would be to build a template and leverage that as you build each job. Something like FastTrack comes to ...

Asked and answered. If you have a question that's a little more specific, please ask it but that one has already been covered.

1. However you like. What do you want to control it - a database? Something within DataStage? For the latter, the simplest is probably to use the supplied "KeyMgtGetNextValue" routine.

2. Sorry, not really sure what you are asking. Can you clarify the question, please?

Still wondering... isn't there an option in the Connector to supply your own SQL when using it as a source?

Was just coming here to post the contents of the File option for the connector. File Use this property to specify the name of a text file that contains the SQL statements to run for each input row. Include the path if the file is not in the project directory. This property is required when the User-...

True... never did ask how many nodes this job is running on. Does it change (get increased) when the rows to process goes up?

All I'm saying is that it is known to work in other stages and thought it was worth a shot in the Teradata Connector. I personally don't have any way to test or confirm that but if you can find a way to put your own custom SQL into the stage, then this syntax may allow it to be read from a file inst...

Knew it did on the Server side of the world, wasn't sure enough about the Parallel side to make that same assertation without access to my docs.

Those two strings look the same to me. :wink: [Andy, Andy, Andy] The forum software removes any 'extra' spaces automatically... you'd need to use tags in order to preserve them in any posted examples. Did you check your documentation for the permutations of the "Trim" function? Don't recal...

I've seen that behaviour in the past where it squirrels away a copy somewhere unexpected. In the Old Days you could right-click on the stage and select 'Grid View' and find it sometimes, don't think that's an option any more. Kind of surprised that it didn't show up in an export of the job. Ah well.

The whole point of the linked conversation is to note that there doesn't need to be a specific option for this to work. Can't tell from what you posted - did you try telling it to use User-Defined / Custom SQL and then put {FILE} and the file's pathname where the SQL would normally go?

Perhaps... there isn't always a specific option to do so but see if this helps.

OK... sounds like you never did but after rebuilding it again it now works? In your shoes I probably would have exported the job and then gone through the dsx with an editor to see if you could find where it was still referencing the old name.

Sounds like another case where you were dependant on a bug in the 8.x release that has since been corrected. Meaning, we've seen similar issues in the past and the solution is to correct your code to match the corrected behaviour of the job. In this case you should be checking for NULL rather than '...

It's too late to remove your AV software if that is the culprit here. One change you could make is to ensure the directory where DataStage is installed is configured in the software as someplace to not scan. Also check whatever quarantine logs your software keeps to see if it has decided to remove a...