DSXchange

Does this need to be a PX job? There's an equivalent to the unconnected lookup on the Server side, hence the question. Not saying there isn't a PX answer to this, but nothing is popping into my head at the moment other than the Server solution.

Out of curiousity, what 8.x version do you have?

If your join key columns are null on one side or the other, then you'll need to look into some flavor of a outer join. An inner join will never return any records where join keys are null.

So, look closer at your job design, see if anything about it can be made more... optimal.

As you've seen, the Director warning default is set per user. There is no global / project level setting for this. However, keep in mind the fact that when you productionalize your jobs, they won't be run by hand from the Director under several different userids, but rather either a single one or fr...

Well... seeing as how your old version had no such built-in function, did your shop perhaps write your own NextSurrogateKey() version? The 8.x version introduces that as an alternative to the Surrogate Key Generator stage.

RCP has nothing to do with 'reducing the total run time' of a job, it's all about creating more 'generic' jobs.

A join stage does a join and you've included more than just join criteria in your specification. This part is the join: D1 left join D2 where D1.C1=D2.C1 and D1.C2=D2.C2 Whereas this part is a filter and something you would need to implement post-join: and D1.C3='' and D1.C4='' Also note that if you...

kanasai167 wrote:what im trying to do is something like this :
Sequential file ---> DB2 UDB API

What happened with this? That should be a perfectly valid job design, what issues did you have when you tried this?

Ah... dang, should have thought of that as a possibility, I've had that happen to me as well. Good job on figuring that out.

Now, that's a different question. They were asking if it was possible, not if it was advisable.

Anything that works with one userid and not with another screams permission issue. What's different about this third job that hangs versus the other two that work?

They can co-exist on the same server if they support an "itag" installtion. Before this, I would have said no but I seem to recall a recent post where it was noted that the 8.5 release does in fact now support an itag install.

Can anyone confirm or deny?

There's no reason you would need to do this in two halves, but you certainly can do it that way if you like. If you are having some kind of a problem getting the data from the flat file into the database, tell us why, what errors or issues you are facing. Simply saying you "can't insert data&qu...

Googling for "SQLSTATE=S1010" makes it seem like a coding error since this is basically a sequencing issue, as in functions are being called out of sequence. No clue if that's a driver issue or an ODBC stage issue but in your shoes I would take this to my official support provider and see ...