DSXchange

Assignment of value is not possible in a Modify stage. You can do it in a Column Generator stage.

How much memory do you have? How much scratchdisk in the nodes on which the Lookup stage operates? How many other stages are there in the job? How long is a piece of string? (Try it and see.)

Then you're going to have to find out where the truncation is occurring. For example, is there a SUBSTRINGS function in the extraction SQL? Is there some point in the job where the field is a VarChar(26)? DataStage will NOT exhibit this behaviour unless it is programmed to do so.

Resist stupid requirements! Write to a new file then delete the old and rename the new. Or rename the old, process it, and delete it when done.

You could certainly adapt that or the BGA widget using the REST API.

Yes. The size of each data type is documented (from memory in Chapter 2 of the Parallel Job Developer's Guide). Add these together for all the columns in the row that you're processing, then multiply by the number of rows in the reference data set.

A DataStage parallel routine is an external function, written in C++, to perform any task that you like but which, for example, it is too tricky to achieve using regular Transformer stage expressions. One example is the ability to replace one substring with another anywhere in a main string, the so-...

For example, you might create a shared container for generic error handling. Within the container additional columns could be added for the job name, the job start time, etc. In parallel jobs, other columns can be made to flow through the shared container without being specified, by employing "...

From Director choose the Project menu, then Print. Within the Print dialog you get a Print to File option.

Yes, but you must use the orchadmin command (dump option) which, in turn, means that the script must set APT_CONFIG_FILE and APT_ORCHHOME at a minimum.

You'd have to ask IBM about that. As a "standard" it seems to me to be almost useless. If your job isn't doing any processing, then it isn't waiting on that and can demand as much I/O as the system can provide. Moving things to different (more) spindles will not reduce the total demand for...

Do you have any good reason for using unbounded VarChar? If not, use VarChar(50) and you will guarantee sufficient storage for all characers in the string.

Include environment variable $OSH_PRINT_SCHEMAS as a job parameter and set it to True. This will show you which schema has field "abcd" defined as an unsigned integer.

Or that it has been changed in Designer and not yet (successfully) compiled.