DSXchange

From what I recall, it doesn't like the space before the AM or PM when doing 12-hour format conversions. Try removing it and rerunning your tests.

Better yet, search the forums for Controller problem: Unhandled failure for the answers you seek. It's only been discussed a bajillion times.

Well... this isn't SynopsisXchange. What are you trying to find out?

There's no such restriction on CPU usage for jobs run by a Sequence job. The O/S decides where everything goes, regardless.

No, the post was primarily in response to Raftsman's post, you just snuck in there at the last minute between 'Previews' and I didn't bother to adjust for it.

The 'ps' was for you, however.

I'd be really careful about giving advice like that - any idea how many processes that rely on the NLS_DATE format being defaulted that you might break with that change? If you have a successful process, why not stick with it? The 'default date format' being discussed is for external processes and ...

Odd... if you Reset the job after it aborts, is there a 'From previous run...' entry in the log? If so, please post the contents.

Need a better idea of exactly what you are doing in your job. One ODBC stage, two output links, each one using User-Defined sql? If you use two separate ODBC stages, do they work fine?

Thanks for posting that, Greg. :D

Would need a better idea of your job design and constraints. I'm also assuming your job fails with the NOWAIT error rather than finishing properly and then other processes get the error attempting to access your table. I would guess that while you may only be sending data down one path, all are firi...

I would typically script a 'checker' like that. Then with all metadata the same, process all files at once by concatenation or via the Filter option. Otherwise, a looping structure can be built to process them one-by-one.

Captain, we appear to be caught inside of some kind of time loop...

Right, was wondering if that was what you meant - by explicitly padding with another character. However, an unpadded string on the other hand..

If you write Varchar(10) "Hello" to an Oracle CHAR(10) you are going to get "Hello " unless you try very hard not to. I'm not sure how you keep a CHAR field from padding with spaces, even by trying very hard. And keep in mind the fact that Toad has an option to automatically 'Trim string data in CH...

There's no 'condition', really - a new file is created whenever the value in the Trigger Column changes. Doesn't matter what the value is or if the column is actually used in the XML or not - when it changes, you get a new file. We use it to keep the files under a certain size - not an exact science...