DSXchange

ml wrote:I think that may be you need to add the ORA home directory in your ..DSEngine/dsenv file.

Only valid advice for UNIX servers, I'm afraid. No such thing on Windows.

Yup, very nice stuff that we use extensively. In the past his KBA Job Control Utilities have been freely available for download from his website and I imagine they still are.

I can verify this - while sql functions work just fine in a column derivation of active stages like an OCI stage, for some reason Job Parameters cannot be part of the mix. Bug or feature I have no idea, but there have been a couple of times I really wanted to do that and have to keep remembering tha...

Don't use this stage, but I vaguely remember you need some portion of MQ Series installed on the DataStage server and it seems to be complaining it can't find it. If you do have it installed, then its path needs to get added to whatever Windows uses to control that - PATH perhaps? On a UNIX server i...

Any such patches would be found on Ascential's eServices website.

Wait, I want to play too... the difference is the number of rows where input.colX is null.

The error message tells you the problem. The transformer stage is actually compiled rather than generated osh script, so a C++ compiler is needed. And it either couldn't find yours or didn't like what it found.

Call it what you will, but the Reporting Assistant is old and does not support Parallel jobs. I believe that "they" expect you to be using MetaStage for that kind of functionality across all job types nowadays.

As Ken noted, the 'dsjob' executable is used to (amongst other things) run jobs from the command line. In addition to the script he linked you to, read the documentation that comes with the product - in particular the Command Line Interface chapter in the Server Job Developer's Guide manual.

Right... so you:

1) Create new project
2) Export everything from old project
3) Import into new project
4) Delete old project

You may want to wait for a day or two after step 3 before doing step 4.

Great... even better.

You won't as it is not a valid option for a hashed file - you'll only find that option there when the link is to a UV stage or an ODBC stage. I'm not going into the gory details as they been posted here ad nauseum, but you can point a UV stage to an existing Hashed file and then you can get the hash...

Doesn't matter what project an export from made from. You can take a full export from any project and import it into any other project, regardless of the project name on either end of the process.

Hashed files do not natively support 'Multiple row result sets'.

kollurianu - it's a Link property. Get to it by right-clicking on the lookup stage from inside the Transformer.

Edit your first post and repost the routine code with Code tags around it so it preserves the indentations and we can actually read it.