DSXchange

You only have access to the rows in the per-record code. Therefore you ought not to be writing your own loop. The variable called y is behaving properly like a static variable (I misunderstood your original requirement); try making it a regular variable.

Do you put the job parameter reference in the Command field or in the Parameters field? Try the latter.

Are you using the same configuration as the one used to create the source Data Set? Set APT_DUMP_SCORE to True and examine the score that is logged, to verify what degree of parallelism is being used.

Rows/sec is not a reliable metric, not least because row sizes vary. Prefer MB/minute.

You're going to have to incorporate some debugging statements in your code to capture the actual values of x and y and record them somewhere so that you can work out what's happening.

Did you search for 81011? It means "host name not valid or server not responding". Verify that the -host option is correct in your dsjob command. The -host option is required unless you are executing dsjob on the DataStage server machine.

Decide upon and rigorously follow a stage and link naming convention.

Your query is going to need at least one parameter marker so that the ODBC driver can understand the logic of the deletion. For example, the job can deliver the value "HK" as the only column in the ODBC stage. Your SQL (largely unchanged) would then look like DELETE From postal_loc_pickdel_test1 WHE...

Not much you can do unless you can diagnose why it happened when it did. For example, are there multi-instance jobs involved one of which might have locked another out? Or is the hashed file write in a shared container in more than one job?

Templates are stored on the client machine. Perhaps the path you have given does not exist on your client machine, or perhaps you lack write permission to that folder. The pathname of the folder in which templates will be stored is prompted for when the client software is installed. It's usually a s...

There is only one @INROWNUM, and only on the stream input link into a Transformer stage.

DSLinkRowCount (or DSGetLinkInfo) can be used on every link in the job.

A DataStage job that selects and inserts will, of itself, employ cursors. You don't have to code them. Just provide the SELECT statement in your "source" or "extract" stage, and the INSERT statement in your "target" stage.

Welcome aboard. :D

The "." means "current directory". If you're executing a DataStage server job, that means the project directory. There's no way around that. You can't use ".".

Use a job parameter instead, for example #FileDir#/myfile.txt

It is a completely different product that happens also to be called DataStage. Indeed, there are four separate products that are called DataStage, with very little overlap between them.

Welcome aboard. :D

Initialize the variable in the "before" code, and use it in the "per record" section. If you must, create a static variable. If that term is unfamiliar, enrol in a class to learn some C.

JOBCONTROL.H is a header file for server routines, not for parallel routines or BuildOps.

Because JOBCONTROL.H is a header file, you do not call it. You include it.

There are many other header files that are relevant to C++ coding, as Arnd has suggested.