DSXchange

Not an 'alias' but yes, a transformer to map the source columns to the target columns. There may be a 'better' stage for this for all I know, but that would work.

pavan.nadiger wrote:An entry for the new column was made in all the relevant stages.

Well... the behaviour suggests that this isn't true. I'd suggest you triple-check your job. And this seems like one of those times when getting another pair of eyes to help would be prudent.

ds_team wrote:Where as when I run the same job sequence on datastage server version 7.5.1.A, it executes successfully.

If that's true and the exact same series of jobs work under 7.5.x but not under 8.x I'd say it's time to call in the calvary. Open a ticket with your Support provider.

There's multiple ways to accomplish this. A job could source from a transformer, pass no rows to a DB stage and simply perform the 'truncate'. I have a multi-instance job that 'truncates any table' with the InvocationID being the table name. Food for thought. And a separate truncate step may very we...

Then you'd be best off contacting your official Support provider, get the official answer for those specific versions.

Perhaps something like this would help.

What is your job design, I'm guessing just the two Oracle stages?

If so, you'll need something between them to map the columns. For example, to map 'a' to 'd' (etc) in your second example.

Sure, any script you can write can be run from the Execute Command stage... but are you sure that's the right approach? With the truncate built into the load job, it can be easily restarted. However, what if your load fails and the solution is to truncate again? Now you've got to find a way to rerun...

Searching for your error message (leaving off the job number, of course) turns up 35 matches of people having similar problems. Some even with solutions, like this one.

DS4DRIVER wrote:Is there a way to Supress Trialing 0's?

Math. Add zero to the field, for instance.

Thanks Ray.

Where / what exactly is this 'testing environment'?

You don't have alot of migration choices - it's pretty much either Version Control or Export/Import.

There's a wealth of information here on getting information out of job logs, all you need to do is search for it. Check the online help for the various DSGetLog related functions, if you want to write a routine, or the Command Line Interface section in the Server Job Developer's Guide if you want to...

tsamui wrote:Instead of using 'ls' use 'ls -m'.

+1. Easiest way to build a delimited list of files for consumption by the Loop stages.

Ok. The original post led me a little astray but I see regardless of scale it's a precision issue. As Ken notes, I've had to switch to the 'string math' functions (and write an 'SMOD' routine) when working with numbers larger than the EXACTNUMERIC setting, which defaults to 15 not 4 or 9 or 7 or 14....