DSXchange

That has to be a bug. Advise your support provider.

I can not read your mind, so can not determine your particular intent. :D However what you say is true if you use "documented effect". As you note, you don't even need to include it in the job as a parameter - simply having it set (and therefore active in the environment) in the project will suffice...

In the Connection branch in putty, what is the terminal type string being sent to the server? This becomes the value of the TERM environment variable in the operating system, from which dssh gets its initial terminal type. Mine is currently "xterm". Your SET.TERM.TYPE needs to match this value. Ther...

You can convert each "~" to "'~'" using an Ereplace() function [search the forum] and concatenate a single quote character to each end of the string. Or, if the source is a file, you could pre-process it with a stream editor such as sed (I note you are on Windows, but assume you have MKS Toolkit or ...

I did read it completely. If you take the output you have in your initial post and process it through a Remove Duplicates stage keyed on Col1 and Col2, retaining the Last record of each group, you will have the answer you desire. Which is precisely what I suggested.

HOW are you trying to use DataStage to convert it to VarChar, and what would you expect to get from this particular string if you did? (And do you have NLS enabled?)

Yes you can. ANY executable can be called from an External Filter stage (subject to the following restriction if it is to work properly). The script must accept (on its standard input) the stream of data from the input link to the External Filter stage, and generate (on its standard output) a stream...

Check permissions on /41bdsadm/Ascential/DataStage/branded_odbc and its subdirectories. You need at least read and execute.

The main difference is how data are stored - in a Data Set internal formats are used, while in a File Set external (human-viewable) formats are used. There are implications from these but you only sought the main difference. When is the interview?

Keep in mind (and in your design) the consequences of parallel execution. Is "the tenth record on each node" what you require, or "the tenth line from a sequential file (being processed sequentially)" what you require?

Yes. The Merge stage allows a separate output ("reject") link for each secondary input.

Is the Merge stage really the one you want? The first time you match a "Y" it is consumed, so the second time you try to find a "Y" there won't be one.

Put logging entries into the Job activities that invoke the Audit sequence so that you can determine whether or not they get triggered. Make sure that the invocation ID is reported. Disable log purge on the Audit sequence. It may be that they do get triggered, but you don't see them. For example, if...

That is an Oracle restriction, not a DataStage restriction. Other database servers have similar (but different) restrictions.

If it's the project that's "from scratch" then you have an excellent opportunity to do things right from the beginning - proper metadata management and control, adherence to in-house naming conventions and, above all, documentation. Grasp the opportunity - you won't often get it. And remember, the s...