DSXchange

Can you post the actual entries that you created?

What exactly was the 'setup' that you performed? For any ODBC connection you'll need an entry in .odbc.ini under $DSENGINE and uvodbc.config under the project directory, have you done so?

All of the documentation installs on every client pc.

Sorry, but you enabled an option that says 'treat errors as non-fatal' and then you're wondering why your normally fatal error was logged as a warning?

Yes, it is due to this that the job is not aborted. Uncheck that option.

dsadm is an 'admin' user for DataStage but is not quite the same as a 'superuser' a.k.a. 'root'. What happens if you do the actual startups as root?

Not directly in a single job, no. Typically, that '1st ODBC' work would be in one job and the results would be landed somewhere, say in USERSTATUS. Then a Sequence job would read the value from user status and assign it to a Job Parameter in a second job where the '2nd ODBC' would use it in its wher...

Exactly. IBM / Ascential provides a utility to do that for you automagically.

Look at the generated .ctl file. I think by default sqlldr handles all strings as CHAR rather than VARCHAR (hence the 256) so you would need to use your own .ctl file and thus a 'manual' bulk load call.

I'd also be curious if this is the first time ever you're attempting to use the stage or you've used it in the past and the issue is just with this connection? For the former, I would wonder if all of the 'configuration procedures' have been followed that the documentation mentions must be done befo...

Perhaps, with a little extrapolation, this can help.

A stage variable is a derived field, built row-by-row in the order listed. They are equivalent to Variable Ports in Informatica, if that helps. Use them to make complex expressions self-documenting, or to perform a complex expression once and then use it in other derivations rather than re-evaluate ...

vinothkumar wrote:Use DateFromDaysSince function.

No such thing in a Server job.

Oddly enough, most questions have been asked before.

An exact search for "will partition despite the preserve-partitioning flag" would have answered this for you.

It will always 'work', it will log a warning in every job run if it thinks you are out of compliance.

So, if you remove the target OCI stage and replace it with a flat file target, the job runs at 12k r/s? Those 'combo actions' can be dog slow for volume loads, why not prequalify the action and run two links - dedicated Insert and Update links?