DSXchange

You left out the actual "sqlcmd" bit on the rewrite.

Well... was working off the "if any one fails" comment. Still, bottom line, use appropriate triggers to control the flow from each job to the Sequencer based on your requirement.

Looks to me like Aggregation, grouping on the first three columns and then min/max on the other two.

Not sure why you felt the need to hijack a completely unrelated thread for this question, regardless here we are all split out on your lonesome. Please take more care next time. Also note that I've had to guess at your particulars regarding version, O/S etc... if any are wrong please edit your post ...

That and make sure the trigger expressions coming into the Sequencer are both set to 'Finished OK'.

Doh... missed that part. I'm surprised you were able to select it from inside the stage, thought it filtered out things of that nature.

I'm going to pipe in here from personal experience and clarify something both Vincent and Ernie have mentioned: DataStage 8.1 has a poor XML Pack that does not handle high data volumes how big are the INDIVIDUAL xml documents. The size of a SINGLE document is the key here. IMHO, Ernie's question is ...

Reset the aborted job and let us know the contents of a "From previous run..." log message, if one shows up.

Typically in that case one would need to check the string for spaces and only 'unpack' it when the value is something other than spaces. I don't think the CFF stage has an option to support that, worst case you'll need to do that in a downstream transformer as noted.

Post the actual, unedited error from your job's log. There should be an Oracle in there somewhere as well, me thinks...

Detail for us your Sequence job... what does it do? What stages does it use? What kind of trigger expressions are you using? etc etc etc

To me that would depend entirely on what exactly you are changing and if you get it 100% correct or not. Minor stuff, maybe, (although we may disagree on just what 'minor' means as I've destroyed a project importing a dsx file with a minor change) but I for one would never consider doing anything li...

... and even then Oracle will turn the empty string back into a NULL when you load it, from what I recall and noted above. Is there some problem with a null in your source being a null in your target as well?

Substring.

That's not a "hard limit", more of a soft one in Server as 15 just happens to be the default value for precision in the configuration file. And "after the decimal point" would be scale not precision.

Have you tried it? Had any problems with it?