DSXchange

My thinking is that the query from DB2 will be cached in a virtual data set, and therefore doing one would be less demanding of memory than doing two. Both outputs from the Copy stage should be able to use the same data set. Beyond that I don't really know what other factors would influence "efficie...

You can make them read-only using Version Control as part of promoting from development into test or production. You can toggle the READONLY value in the DS_ROUTINES table (search for this one). If it's a C routine, you can use operating system permissions to protect the source and/or compiled versi...

Migrating DataStage is much more than moving things from one disk to another, just as migrating any database product is much more than moving things. Now that you've done so, there are quite a few repairs that you will need to do. It might be worth hiring in a competent consultant. Needed repairs in...

Keep counts of Y and Z records. Detect these values when an X record shows up. If either is zero, reject the file (stop the job?).
Otherwise reset counts to 0 for X record after making the test (for example in another stage variable).

Be more precise with your requirement. What, exactly, is your required output for each line processed? For example, for each X record, do you need the number of preceding Y and Z records (separate counts or combined count)? What is the output for Y and Z records? The solution will certainly make use...

Omitting error handling for clarity:

If there's no match to the hashed file, then all columns returned from the reference input link are set to NULL by DataStage. You can capture the NULL by direct reference to the input link; otherwise there's no value to capture, so you might (for example) set a default value (999) in the output colu...

This behaviour also occurred in 5.2. When compiling a job that uses a shared container, the compiler attempts to set a lock on each referenced shared container (Designer already has a lock on the job), so that they are not changed during compilation. The only solution is a management one; you have t...

You didn't mention FTP in your original post! The solution I gave was one that also didn't assume being embedded in FTP script.

I would have to give it a great deal of thought. In the meantime, try an FTP stage, and run find as a "before" command. You may be able to get that to work.

The "end process" from Director and from DS.TOOLS, and the "UniVerse" command (MASTER OFF) all send a signal ( kill -15 ). If the process is not using any CPU cycles (for example if it is sleeping) it will not be able to service this signal. This is why "log out process" sometimes appears not to wor...

Hey, if the information you need is already being recorded in "QA tables" why not retrieve it from there? Chances are that the pages needed will still be cached.

You don't seem to understand the statements. OPENSEQ doesn't create anything. CREATE or WEOFSEQ creates an empty file (one with zero columns). WRITESEQ writes an entire line, including the line terminator. Therefore you need to construct the entire line, including delimiter characters or space paddi...

No. Because you haven't specified which column(s) specify the file and which are data to be written to the file. You need to pass all needed columns as arguments, naturally. If there's only one column being used to differentiate output files, the only change is to add the other columns to the WriteS...

I don't think anything's moved, otherwise /.uvhome would be wrong, and a different message would have been generated. However, if the permissions are wrong or the installation is incomplete the indicated error message could have occurred.

And thanks to you and your associates for posting such a detailed answer, which means no-one else need ever be unable to solve this particular problem. :D