DSXchange

That's the beauty of the design; you're NOT changing the structure! An I-descriptor is metadata describing a virtual field (think of it as a view or as a trigger - this will be clear when you see the technique). Indeed, if your customer is running UniData 6.0 you can use triggers directly. And if th...

When they pay me. Yet another 'funny' story - they recently replaced their old, 'outdated' mostly manual invoice processing system with a new spanky web-based one. There is a lovely Portal with an automagic process behind the scenes to make all of our lives easier and hooks to Brio reports so we ca...

OK, some diagnosis. (I take it your sojourn through file dictionaries found nothing. There are some UniVerse files that don't have dictionaries; you should not be concerned about these.) The message is generated by UniVerse/SQL, which means that the alleged association (CCIS_UV_TR_CCIS_ITEM) is bein...

Must've required some Deep Thought to figure that out!

Is the DB2 instance on the same machine as the DataStage server? Put another way, are you performing these lookups over a network? Think about what this involves; assemble a packet with the key expression, send it to the database server, await its response, unpack the packet in which it was returned...

The Aggregator stage starts with an in-memory buffer of 8KB, which doubles each time it becomes full until it reaches 0.5MB. The next multiplier is 8, so from 0.5MB it jumps to 4MB. These sizes and thresholds can be adjusted using option 6 from the DS.TOOLS menu. Pre-sorting the data on the grouping...

From what I understand you want to have the initial load done on 31 Dec (which it already is), and incremental loads from that point on. Initial loads of master (or dimension) tables generate new surrogate keys, which you can use DataStage or a sequence or serial data type in the database server to ...

Yes it would.
Just as useful would be the detailed event message from the job log, so we can verify the actual names in your code. If there is some privacy issue with this please say so, and I may be able to take it off line.

This looks much more like a job for a Merge stage. If this is not installed in your DataStage project, you will need to install it using the DataStage package installer (find it via the Start menu, under Ascential DataStage). The package itself is on your DataStage CD, in the Packages folder. Also i...

It has been suggested to us that Creating our hash files with the minumum modulus set to the current size listed from ANALYZE.FILE will improve our performance since the file will not need to be resized as much as it otherwise would. (it grows daily) This seems to have been working for us, but now ...

kcbland wrote:Visit after a strong cup of coffee:
http://www.ralphkimball.com/
http://www.inmoncif.com/

Even two cups wasn't enough!

Hi, I actually ment to ask whethere the "if Not(hJob)" is good enough to check for DSAttach(...) success. Nope, read the BASIC manual. It states that any value can be returned. Your logic assumes that a 0 return code is an invalid handle, which the manual says is not true. This is a gripe I have wi...

Not really, unless you create a Routine to do it. Algorithm design is: 1. open VOC file 2. select local file pointers (use DSExecute and SELECTFL) 3. for each file selected: (a) open its dictionary (b) attempt to read an item with the sought name (c) display message i...

When you use the term "SQL" do you mean the query language or the Microsoft product "SQL Server"? I appreciate that English is not your first language. If you mean the latter, are you asking how to connect from your UNIX server to an instance of SQL Server running on a Windows platform? This questio...

You should have all the manuals on your DataStage client CD. The base set of manuals (including Parallel Job Developer's Guide) installs with the client software, in a folder called Docs - unless you do a custom install omitting them.
The remainder are in the Packages folder.