DSXchange

It's probably best to use DirectoryAdmin.sh to effect the change.
http://pic.dhe.ibm.com/infocenter/iisin ... tcred.html

Try it without using the namespace references but, instead, using the explicit URLs that belong to each namespace. I have found this to work in the past. Sometimes, it seems, the WebServices Transformer stage simply does not work properly with namespace references when the header is customised.

The compound should be the business key, though a more efficient fact table would be constructed by amalgamating these into a single column, perhaps with delimiter characters. The Surrogate Key must be an integer of some kind, usually int64 (or even uint64) to allow for growth.

Optimization changes never affect explicitly-set partitioning.

Optimization will only insert partitioners where there are none.

None of your approaches achieves the required result of avoiding CoSort licensing, as you're still planning to invoke the CoSort scripts. Why not try implementing one of them using a Sort stage, perhaps increasing the available memory per node when performing the sort? Make sure your data are partit...

You wouldn't have missed much, so I unprotected it.

Yes. The fact that multiple columns are marked as Key in the metadata drives this behaviour correctly.

sudha03_vpr wrote:@OUTROWNUM can be used in transformer to get the number of records coming out of the link.

... per node

ISO8859-x (for any x) includes "ASCII" characters.

Running out of memory during compilation. Is your system under-configured?

Can we sell you a Business Glossary ?

Thank you for posting that. I'm sure it will get someone out of that particular hole more quickly in future.

Have you moved the "table definition" into the project?

WebService Client stage is one-way interaction. You need WebService Transfomer stage if you need request-response interaction. As a general rule the default header that gets built is inadequate, particularly if it needs to include security information, so that you're better off building your own in ...

vivekgadwal wrote:Also, you can check and see if "Multiple readers per node" is beneficial for you or not.

Probably not. The problem occurs when writing a file.