DSXchange

I can't think of an easy direct way to do this in a PX job offhand. It isn't difficult to do with a Server job. How big is your data volume?

I use this method all the time with RCP and generic jobs. Often I want to process tables whose key column have different names so I pass the key in as a parameter, use a modify stage to define a "known" key name, and can then process that constructed column through whatever manipulations a...

Is this a new installation or has DataStage worked before on this system. If it has worked before, what changes were made between the time it worked and now?

Since transform loops, which would solve your problem, are only available in later releases that solution cannot be used. What is the maximum number that can be in Column B? If it is a known, and small, number {n} you could make a transform stage with {n} outputs each coupled to a constraint that wo...

Might it be possible that you installed some other software (or had software installed/updated automatically) on your client PC? This might have overwritten DLLs and thus caused your problems. A client re-installation would solve your problem if that were the cause.

I would recommend using a Modify stage to create a new column (call it TempKeyCol) from the real column name as given in the parameter #KEYCOL#, then have your SQL do

"RGGEM" is not a reserved word for a variable so should cause an issue. What does your constraint look like?

That sucks - particularly if the company doesn't respond and claim to hold copyright. A quick overview of the other articles seem to confirm your suspicions. They are certainly on my blacklist!

While not thrilling news, I'm glad they confirmed what you've already found out. I'm glad I'm using BuildOps here - the one I ran today processed 114 Million records, each about 1Kb wide

"IIRC" -> If I Recall Correctly

Doing runstats on DB2 is always helpful, but in your case that would optimize access times and queries on your repository, this could improve response times for viewing logs in the director or saving/compiling jobs in the designer, but won't make your jobs run faster (unless each job writes tens of ...

Another thing you might try is to make a copy of the job you wrote for 'abcdef' and replace the DB2 stage with a row generator and see if you still get the prepended junk characters.

What sequential file options are you using? Are you adding a row number? Anything out of the ordinary? Does a PEEK stage show these extra characters?

If, in DB2, you "select len(Col1) FROM YourTable WHERE Col1 LIKE '%KNAGE228495356jhg%'" do you get 17 or 20?

What about if you do the same in DataStage?