DSXchange

If you're coming all the way up from 4.2, you've got an awful lot to "get used to"! Most of it is even good.

It means pretty much exactly what it says: You are allowing null key values to be written to your target hash. This is a Bad Thing.

Unfornutaley there is no stage in datastage to remove duplicates automatically. Actually, I believe there is on the PX side. No such luck on the Server side. Warning: Opinion. Using the "insert then update" update action or its twin brother allow ETL jobs to get "sloppy" and can hide a multitude of...

Yes I do have values set for display on the columns tab. In one job I've got 18 columns with displays ranging from 1 to 30 characters. In the data browser the biggest width it gives me is 6 characters. It looks just like an attempt to fit all the columns to a page which is why I thought it might be...

The 'trick' is to not View the log first, but to clear it from the Status view. It will still take a little time for the dialog to come up, but it is only counting them (looking for the most recent record) rather than trying to suck all of them up to your client. You are taking advantage of the Auto...

Starting with 7.5 this is a 'non-issue'. DSjob is now installed with DS client installation -------------------------------------------------- The dsjob command-line utility is now installed as part of the DS client installation. This will enable customers in Unix server environments to control jobs...

In what context? I would think that 'broken' equates to 'corrupted', but I'd be curious where you saw this phrase.

Since we're throwing out ideas, if you are using Oracle 9i you could access the flat file as 'external table' via the OCI stage. Again, way slower than simply loading into a Hash.

Simple - Load it into a Hash file.

Search is your friend because, as you may imagine, this topic has been discussed quite a bit on the forums.

For starters, here is one possible starting point.

Yes, you've got it right... that's exactly what the 'sort' option is for. Plus you seem to understand the fact that you are not asking for it to be sorted, you are asserting the sort order that the Aggregator will see the data coming in with. If you have the sort option checked off and it didn't mat...

The fact that projects export with an extension of .dsx has no bearing on what you need to do to get them scheduled. And if you are running a Windows server as your post indicates, then korn shell scripts aren't going to be a whole lot of help, either. In any case, Ketfos has you on the right track ...

From a Client standpoint you can certainly do it. Just make sure you install the new client into a different directory than your current client is in. Then, depending on the exact version of 7.x you are running, there is either an 'unsupported' client switcher (7.0, 7.1) or a new 'supported' one (7....

Thanks for the reply. I guess I'll find out when I actually get a chance to install (and test) this!

If you've been diligent about recording dependencies (this is why the Dependencies tab is there), then Version Control can pick up the dependent objects. Actually, unless something has changed and I haven't noticed, VC does not need the Dependencies tab to find dependant objects. I'm not saying it ...