Search found 53125 matches

by ray.wurlod
Wed Nov 30, 2005 12:51 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Automating Load Process
Replies: 2
Views: 967

Give large amounts of money to a competent consultant. Magic happens.
by ray.wurlod
Wed Nov 30, 2005 12:50 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Conditional lookup using hased files
Replies: 13
Views: 3878

Conditional lookups are fully supported in the parallel Lookup stage. Who was asleep during class? A server job solution to implement true conditional lookups would require at least two Transformer stages. You can do it with one Transformer stage and two reference input links, but both lookups will ...
by ray.wurlod
Wed Nov 30, 2005 12:46 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: DB2 UDB API Connection
Replies: 7
Views: 2287

I believe you will need to have DB2 Connect installed to be able to use the DB2 UDB API from a UNIX machine to access tables on an AS/400 machine.
by ray.wurlod
Wed Nov 30, 2005 12:45 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: How to connect to DataStage for test automation
Replies: 11
Views: 5741

dsobjects.dll is already a COM object. One of the objects exposed therein is a Subroutine object which can, of course, return values. But it's probably a cumbersome approach, and you need to establish a connection to DataStage using a Session object. "They" don't provide any documentation on this, ...
by ray.wurlod
Wed Nov 30, 2005 12:42 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: NLS question re: Spanish characters
Replies: 13
Views: 5482

"Conversion in DS is none" really means that, if you have NLS enabled, the mapping within DataStage should be set to the pseudo-map NONE.

If you don't have NLS enabled, the point is moot.
by ray.wurlod
Wed Nov 30, 2005 12:39 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Error while accessing unidata db through unidata stage
Replies: 6
Views: 1724

The only thing I can think of is either to clean up the data in UniData and severely castigate the proponents of keys with marks in them, or to include an additional piece to your WITH phrase in the UniData stage to eliminate keys with marks, such as

Code: Select all

@ID NE @ID CONV "MCP"
by ray.wurlod
Wed Nov 30, 2005 12:35 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Adding description to jobs
Replies: 8
Views: 2767

Descriptions should always go in first. "Later" never comes. I will shortly post a brief essay on the process of job design. It begins without any software (yes - planning!). The very first thing to go into a job design is to add short and long descriptions. The long description is probably a lift f...
by ray.wurlod
Wed Nov 30, 2005 12:31 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: DS PX - cluster
Replies: 4
Views: 1773

DS EE is cluster-aware. You do have to have the osh stuff on each machine, but you do not need the complete DataStage server on each machine (contrary to anything a sales person seeking extra commission may claim!). Player processes on separate machines communicate via TCP sockets, as do section lea...
by ray.wurlod
Wed Nov 30, 2005 12:29 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Job execution problem after C++ installation
Replies: 5
Views: 1568

You need to have installed C++ compiler ahead of DataStage. The quick fix is to re-install the DataStage server. This will not affect any existing jobs. (It will "upgrade" them to the same release!)
by ray.wurlod
Wed Nov 30, 2005 12:27 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: A list of Questions
Replies: 5
Views: 3213

Most of those questions can be answered in Powerpoint - after all, everything works in Powerpoint. However, as the others have also noted, there are large, complex and, at times, subjective factors that would come into play. Perhaps you could ask the client to be more specific with the questions. Ot...
by ray.wurlod
Wed Nov 30, 2005 12:23 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Portability between Unix OS
Replies: 7
Views: 2151

Beware of compiler compatibility issues - recompile everything with a Transformer stage, as Vincent noted, but also any BuildOps and custom parallel routines. You may also need different compiler and linker options.
by ray.wurlod
Wed Nov 30, 2005 12:21 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: DS Warning
Replies: 4
Views: 1819

Read the messages carefully - they tell you precisely what the problem is. You've set the null value property to "" (length zero), which is its default value, but you've specified the data type as string[11] so that DataStage is demanding that your null value be an eleven character string. The other...
by ray.wurlod
Wed Nov 30, 2005 12:19 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Error in Basic Transformer
Replies: 5
Views: 1444

The BASIC Transformer stage is, as Arnd suggested, quite robust, though it incurs a performance penalty as data need to be converted to the "typeless" form used by BASIC and back again. If you created a transformation that yielded a data type that was not legitimate on return, you would see a failur...
by ray.wurlod
Wed Nov 30, 2005 12:13 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: BuildOp stage to call unix script
Replies: 1
Views: 682

You don't need a BuildOp. Check out other stage types, such as External Filter stage.
by ray.wurlod
Wed Nov 30, 2005 12:12 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: MQ Series Trigger
Replies: 4
Views: 2035

Unless you create an "always running" DataStage job (or job sequence) that listens for the particular message you do not have any mechanism within DataStage to do what you request. You could create your own listener external to DataStage and have that trigger a DataStage job via the command line int...