DSXchange

Is the job in a runnable state (for example, not Aborted, Stopped or Crashed)?

When a static hashed file is created you have to specify the size and number of groups (pages). If the volume of data is likely to change these values may become outdated. You need to monitor the level of overflowed groups and resize when necessary. This is managed automatically in dynamic hashed fi...

Merge is a "master and updates" model. Would a Funnel stage fuflil your requirement more exactly?

This appears to be on the correct forum; it is an enhancement request for a ReadMyMind stage.

What about simply targetfield:nullable integer=sourcefield ?

Don't forget the user requirements which, if you don't get signed agreement, will change just before you are ready to go into production. Even if you have signed agreement they'll still change, but now you are in a proper software development life cycle and can oblige them to submit a formal change ...

Whenever you like, provided you don't mind incurring the additional monitoring and maintenance overheads. It's defined on the basis of "file type" (the "create file" options in the Hashed File stage); type 30 is a dynamic hashed file; types 2-18 are static hashed files. The best candidates for stati...

It might be that, it might be something external that has changed, it might be, as your colleague suggests, a network congestion issue. Ot it might be locking in the database, or all kinds of things. Address only one at a time as you investigate.

Doing each test in its own stage will work I think, I will probably write to a hash file so all repeated items will overwrite itself. Are there anydisadvantages I should note if I do it this way instead of bunching all the checks within one transformer stage? My recommendation was separate stage va...

And presumably you agree?

With NLS enabled, data (including design information, etc.) while within DataStage are stored in an idiosyncratic UTF-8 encoding of Unicode (the encoding is called UV-UTF8). With NLS not installed, data while within DataStage are stored as "8-bit ASCII" (also known as ISO8859-1). With NLS installed ...

Are you asking about Type 1 SCD, Type 2 SCD, Type 1/2 SCD or some other variant? The explanation is different in all cases.

I suggest you find a copy of The Data Warehouse Lifecycle and Toolkit by Ralph Kimball and others, in which these concepts are explained in pellucid fashion.

handle$list is a variable that contains a list of job identifiers (job handles, if you like) that the job sequence is monitoring. Since your job sequence does not contain any Job activities, it appears that this variable has not been initialized. This is definitely a bug, and should be reported via ...

One uses a key to identify things uniquely. In database tables, the key identifies the row uniquely. A "natural" key is the key used by the business for this purpose, though they are sometimes quite unnatural! See other post about surrogate key.

A surrogate key is a totally artificial key value that guarantees uniqueness but carries no meaning whatsoever. Bus route numbers are an example.