DSXchange

You guess? Ironic question, but did you test this?

Then you'll need to do it "after job". Write to a static name and add the timestamp after the job finishes.

Sorry, but why are you using UserStatus to "return the success or failure of the job"? Why not just check the job status itself directly? Is this more of a "logical" failure rather than a true abort/warnigs you are checking for?

No, that's not the cause. I routinely use a Sequencer as the end point in a job flow so there's a trigger to check on the last Job Activity ("automatically handle") and there's never a "did not exit" message in the log since there's no exiting link. I think the OP only gave us the "meat" of the desi...

No, not directly though I suppose you could read the file and then pass the contents in as a Job Parameter. As to the format, it depends on your mail server - it may need whitespace, a comma, semi-colon, etc. Write a small test job to check.

True, in looking back I typically only exit the loop When Something Goes Horribly Wrong so don't mind the "did not complete" message. However, the Sequencer message confuseth me a little, I thought that would only being output for Sequencer objects set to "All" when some but not all conditions were ...

I haven't had any issues "jumping out" that I recall, i.e. exiting a loop early. Just don't try to jump back in.

Exactly, and all of which are officially "unsupported".

Just XML Output, I would think. What database? What is the target field's datatype?

Not directly, no. I suppose you could put the logic in a proc and conditionally execute it inside the proc based on the value of the parameter passed to it, but whatever is in the before-sql area will always run.

Unfortunately, I don't "do Windows" so to speak, so don't know all the gotchas for setting up Oracle connectivity there. I'm sure others do, however. I'm guessing you need the Windows equivalent of the "shared library path" setup, something which is a separate environment variable under UNIX but I'm...

#1 sets the value for DataStage jobs when they run. #2 sets it for everything all the time. In both cases you should have added the Oracle information to whatever existed there, not replaced what was there with it. And this is on the DataStage Server, yes? Not your client PC I assume. You don't need...

Which "7.5" version do you have? Literally the first 7.5 release or one of the later minor releases like 7.5.1A or 7.5.2 or ?

It is an optional plug-in, perhaps whomever installed DS chose not to include it for some reason.

Seems like you should be thanking Arnd as well.

Also, I have to specify PATH as ORACLE_HOME/bin directory. I did all that, but no luck. Where exactly did you set this? And you added this to your current PATH statement, yes? You'd either need to spell out the complete path or use the Windows substitution characters for that to work correctly: %OR...