DSXchange

Performance statistics on Designer are only a "nice to have" anyway. You can always get the link row counts from the job log, or programmatically once the job has finished processing rows.

Welcome aboard. :D Whatever column definitions is apposite to the table must be the same in your job design. It's differences that cause this message to be logged. Change your Columns grid so that this column is defined as Numeric (or Decimal) with precision 10 and scale 4. Currently you have it def...

It's actually a general DataStage limitation. The upper limit on the number of links connected to a stage (unless otherwise further limited) is 127. This used to be a documented limit and used also to be visible by examining the Stage Type properties in the Repository; it's still an actual limit but...

Welcome aboard. :D

You can do it with DB2 stage, but will need to have DB2 Connect software installed on the DataStage server machine.

You can do it with ODBC stage, but will need an SQL bridge of some kind or DB2 Connect software.

Also check out IBM's Information Integrator software.

Welcome aboard. :D

There is necessarily an overhead in calling stored procedures. So you try not to; manage the computation within DataStage using local variables and expressions, and you avoid the latency.

You don't use checkpoints. DataStage does, when a job sequence is re-run that failed earlier. All you do is check a box to indicate that you would like this functionality to be compiled in to the job sequence. The only other thing of which to be aware is the "do not checkpoint me" check box in acti...

Export doesn't delete anything - it's just another way to take a backup. Once you've taken a backup - or two - you actually need to start deleting things. And therefore you have to find out which things you want to delete. You can start by getting the job numbers from DS_JOBS. SELECT NAME,JOBNO FROM...

Change NLSMODE in uvconfig to 0 (with uvregen, etc) and prove that there's nothing else obstructing startup of uv.

If all that is OK reinstate NLSMODE then try the NLS button in the Administrator client. Does that work OK?

What error messages do you get when trying to restart DataStage? These "memory offset" parameters are quite sensitive; you are not permitted to generate any shared memory segment larger than the system can deliver; this in turn is configured by UNIX kernel parameters such as shmmax and shmmni. Nothi...

Sounds like the upgrade has lost the files on the server machine in which the permissions are stored. These are .operator.adm and .developer.adm
Find them on backup media and reinstate and magic may just happen.

Find other messages in the log file that tell you why the abort was raised in the hashed file stage.

Aborts in hashed file stages are fairly rare; typically resulting from design errors such as specifying a non-existent hashed file name or directory.

I misunderstood your requirement. The row splitter splits one column into multiple columns. It supports only one output link.

You need a Transformer stage.

THEN IF is always faster than AND (you bail out of the test as soon as any condition is not met). Ordering by likelihood of failure will also help that. Doing it all in a routine add negligible overhead. CASE compiles to exactly the same object code as the equivalent nested IF, so if you think it's ...

This post is very similar to that post - but you can't use Convert() because Windows uses more than one character as its line termination.

Create a stage variable called svCRLF as described in that post to gain some efficiency.

If you're using 7.5 click on the icon on the toolbar that looks like Internet Explorer's icon. This will generate full documentation for the job design, including all derivations.